In this paper, we investigate code-integrated reasoning (CIR), where models generate code when necessary and integrate feedback by executing it through a code interpreter. To acquire this capability, models must learn when and how to use external code tools effectively, which is supported by tool-augmented reinforcement learning (RL). Despite its benefits, tool-augmented RL can still suffer from potential instability in the learning dynamics. In light of this challenge, we present a systematic approach ETIR (Effective TIR) to improving the training effectiveness and stability of tool-augmented RL for code-integrated reasoning. Specifically, we develop enhanced training strategies that balance exploration and stability, progressively building tool-use capabilities while improving reasoning performance. Through extensive experiments on five mainstream mathematical reasoning benchmarks, our model demonstrates significant performance improvements over multiple competitive baselines. Furthermore, we conduct an in-depth analysis of the mechanism of code-integrated reasoning, revealing several key insights, such as the extension of model’s capability boundaries and the simultaneous improvement of reasoning efficiency through code integration. These findings underscore the potential of code-integrated reasoning as a scalable paradigm for advancing robust and efficient language model reasoning.

Balancing exploration and exploitation is a central goal in reinforcement learning (RL). Despite recent advances in enhancing language model (LM) reasoning, most methods lean toward exploitation, and increasingly encounter performance plateaus. In this work, we revisit entropy -- a signal of exploration in RL -- and examine its relationship to exploratory reasoning in LMs. Through empirical analysis, we uncover positive correlations between high-entropy regions and three types of exploratory reasoning actions: (1) pivotal tokens that determine or connect logical steps, (2) reflective actions such as self-verification and correction, and (3) rare behaviors under-explored by the base LMs. Motivated by this, we introduce a minimal modification to standard RL with only one line of code: augmenting the advantage function with an entropy-based term. Unlike traditional maximum-entropy methods which encourage exploration by promoting uncertainty, we encourage exploration by promoting deeper and longer reasoning chains. Notably, our method achieves significant gains on the Pass@K metric -- an upper-bound estimator of LM reasoning capabilities -- even when evaluated with extremely large K values, pushing the boundaries of LM reasoning.

As scaling up training data has significantly improved the general multimodal capabilities of Large Vision-Language Models (LVLMs), they still suffer from the hallucination issue, generating text that is inconsistent with the visual input. This phenomenon motivates us to systematically investigate the role of training data in hallucination. We introduce a new benchmark, POPEv2, which consists of counterfactual images collected from the training data of LVLMs with certain objects masked. Through comprehensive evaluation on POPEv2, we find that current LVLMs suffer from training bias: they fail to fully leverage their training data and hallucinate more frequently on images seen during training. Specifically, they perform poorly on counterfactual images, often incorrectly answering “Yes” to questions about masked objects. To understand this issue, we conduct probing experiments on the models’ internal components, revealing that this training bias is primarily located in the language modeling (LM) head, which fails to correctly translate accurate visual representations into textual outputs. Based on these findings, we propose Obliviate, an efficient and lightweight unlearning method designed to mitigate object hallucination via training bias unlearning. Obliviate identifies the discrepancy between ground-truth labels and model outputs on the training data as a proxy for bias and adopts a parameter- and data-efficient fine-tuning strategy that only updates the LM head. Extensive experiments demonstrate the effectiveness of our approach. While only reusing the training data and updating approximately 2% of the parameters, Obliviate significantly reduces hallucination across both discriminative and generative tasks. Furthermore, it demonstrates strong scalability with respect to both model size (2B to 72B) and training data volume, and exhibits promising generalization to hallucination types beyond object-level hallucination.

With the rapid advancement of large language models (LLMs), retrieval-augmented generation (RAG) has emerged as a critical approach to supplement the inherent knowledge limitations of LLMs. However, due to the typically large volume of retrieved information, RAG tends to operate with long context lengths. From the perspective of entropy engineering, we identify unconstrained entropy growth and attention dilution due to long retrieval context as significant factors affecting RAG performance. In this paper, we propose the balanced entropy-engineered RAG (BEE-RAG) framework, which improves the adaptability of RAG systems to varying context lengths through the principle of entropy invariance. By leveraging balanced context entropy to reformulate attention dynamics, BEE-RAG separates attention sensitivity from context length, ensuring a stable entropy level. Building upon this, we introduce a zero-shot inference strategy for multi-importance estimation and a parameter-efficient adaptive fine-tuning mechanism to obtain the optimal balancing factor for different settings. Extensive experiments across multiple RAG tasks demonstrate the effectiveness of BEE-RAG.

Large Language Models (LLMs) excel in reasoning and generation across domains, but still struggle with identifying and diagnosing complex errors. This stems mainly from training objectives that prioritize correct answers, limiting exposure to and learning from errors. While recent studies have begun to address this by introducing error signals, most rely on shallow, static errors, restricting improvement in deep diagnostic ability. To overcome this, we propose Hide and Seek Game (HSG), a dynamic adversarial framework for error generation and diagnosis, and evaluate it on mathematical problem-solving. HSG involves two adversarial roles: Sneaky, which hides by generating subtle, deceptive reasoning errors, and Diagnosis, which seeks to accurately detect them. Through adversarial co-evolution, both error stealth and diagnostic precision are enhanced. Experiments on three mathematical reasoning datasets demonstrate that HSG significantly boosts error diagnosis, achieving 16.8%-31.4% higher accuracy than baselines like GPT-4o. We also release a challenging dataset of deceptive errors and diagnostic annotations as a benchmark for future research.

Diffusion large language models (dLLMs) offer an efficient alternative to autoregressive models through parallel decoding, yet existing post-training methods largely rely on random masking strategies that overlook intrinsic token dependencies.In this work, we present an empirical analysis of attention in dLLMs and show that tokens attending more strongly to unmasked context exhibit greater generation stability and play a critical role in reasoning.Motivated by these findings, we propose AGDO, an attention-guided denoising and optimization framework that aligns both training and optimization with attention-derived dependencies.AGDO determines the denoising order based on attention structure and emphasizes attentioncritical tokens during supervised fine-tuning and reinforcement learning.Experiments on mathematical and coding benchmarks demonstrate that AGDO consistently improves reasoning performance, outperforming state-of-theart post-training methods for dLLMs.

Yifan Li, YuKai Gu, Yingqian Min, Zikang Liu, Yifan Du, Kun Zhou, Min Yang, Xin Zhao, Minghui Qiu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

The rapid advancement of large reasoning models has saturated existing math benchmarks, underscoring the urgent need for more challenging evaluation frameworks.To address this, we introduce OlymMATH, a rigorously curated, Olympiad-level math benchmark comprising 350 problems, each with parallel English and Chinese versions.OlymMATH is the first benchmark to unify dual evaluation paradigms within a single suite: (1) natural language evaluation through OlymMATH-EASY and OlymMATH-HARD, comprising 200 computational problems with numerical answers for objective rule-based assessment, and (2) formal verification through OlymMATH-LEAN, offering 150 problems formalized in Lean 4 for rigorous process-level evaluation.All problems are manually sourced from printed publications to minimize data contamination, verified by experts, and span four core domains.Extensive experiments reveal the benchmark's significant challenge, and our analysis also uncovers consistent performance gaps between languages and identifies cases where models employ heuristic "guessing" rather than rigorous reasoning.To further support community research, we release 582k+ reasoning trajectories, a visualization tool, and expert solutions at https://github.com/RUCAIBox/OlymMATH.

Xinyu Tang, Yuliang Zhan, Zhixun Li, Xin Zhao, Zhenduo Zhang, Zujie Wen, Zhiqiang Zhang, Jun Zhou. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Long-form question answering (LFQA) requires open-ended long-form responses that synthesize coherent, factually grounded content from multi-source evidence.This makes reinforcement learning (RL) reward design critical.The reward must be verifiable for faithful grounding and stable optimization.However, many standard rewards assume a unique target with an exact-match notion of correctness, which fits short-form QA and math but breaks in LFQA.As a result, current RAG systems still lack verifiable reward mechanisms, yielding unstable feedback signals and suboptimal optimization outcomes.We propose RioRAG, a framework for reinforced verifiable informativeness optimization.First, it defines informativeness as a measurable and externally verifiable objective for RL.Second, RioRAG uses nugget-centric verification with cross-source checks to enable self-evolution of smaller LLMs and to provide denser, actiondiscriminative rewards that mitigate reward sparsity and stabilize optimization.This formulation avoids handcrafted supervision for the policy model and strong teacher-model distillation, relying instead on externally verifiable feedback.Experiments on LongFact and RAGChecker show that RioRAG achieves higher factual recall and faithfulness, establishing verifiable reward modeling as a foundation for trustworthy long-form RAG.Our codes are available at https://github.com/RUCAIBox/ RioRAG.

The rapid advancement of large language models (LLMs) has opened new possibilities for understanding users in search and recommendation. While traditional behavior-based or feature-driven user models rely primarily on explicit interactions or handcrafted representations, LLMs introduce a fundamentally different paradigm: LLM-powered user profiling, where user preferences, intents, and contextual attributes can be extracted, summarized, or reasoned about directly through natural language. This shift unlocks powerful new paths to achieve personalization but also raises pressing questions related to modeling fidelity, temporal dynamics, evaluation methodology, privacy, and responsible deployment. The LLM-UP workshop aims to bring together researchers and practitioners to systematize emerging progress in LLM-powered user profiling, identify open challenges, and explore opportunities for integrating such techniques into search and recommendation pipelines. The LLM-UP workshop adopts an interactive structure featuring lightning talks, panel discussions, and paper presentations to foster active engagement, cross-disciplinary dialogue, and community-driven agenda setting for this rapidly evolving field.

Recently, generative recommendation has emerged as a promising paradigm, attracting significant research attention. The basic framework involves an item tokenizer, which represents each item as a sequence of codes serving as its identifier, and a generative recommender that predicts the next item by autoregressively generating the target item identifier. However, in existing methods, both the tokenizer and the recommender are typically domain-specific, limiting their ability for effective transfer or adaptation to new domains.

Click-through rate (CTR) prediction plays a pivotal role in online advertising and recommender systems. Despite notable progress in modeling user preferences from historical behaviors, two key challenges persist. First, exsiting discriminative paradigms focus on matching candidates to user history, often overfitting to historically dominant features and failing to adapt to rapid interest shifts. Second, a critical information chasm emerges from the point-wise ranking paradigm. By scoring each candidate in isolation, CTR models discard the rich contextual signal implied by the recalled set as a whole, leading to a misalignment where long-term preferences often override the user's immediate, evolving intent. To address these issues, we propose GenCI, a generative user intent framework that leverages semantic interest cohorts to model dynamic user preferences for CTR prediction. The framework first employs a generative model, trained with a next-item prediction (NTP) objective, to proactively produce candidate interest cohorts. These cohorts serve as explicit, candidate-agnostic representations of a user's immediate intent. A hierarchical candidate-aware network then injects this rich contextual signal into the ranking stage, refining them with cross-attention to align with both user history and the target item. The entire model is trained end-to-end, creating a more aligned and effective CTR prediction pipeline. Extensive experiments on three widely used datasets demonstrate the effectiveness of our approach.

Abstract The rapid evolution of large language models (LLMs) has driven a transformative shift in artificial intelligence (AI), reshaping both research paradigms and practical applications. Distinguished from their predecessors by unprecedented scale and advanced capabilities, LLMs necessitate new frameworks for understanding their development, behavior, and societal impact. This survey systematically reviews recent advancements in LLM techniques across four key dimensions: (1) pre-training methodologies, which establish core model capabilities through large-scale self-supervised training, architectural innovations, and data curation strategies; (2) post-training techniques, including supervised fine-tuning and reinforcement learning, which adapt foundational models to downstream tasks and enhance their alignment and safety; (3) utilization strategies, such as in-context learning, prompt engineering, and agentic reasoning, that optimize real-world deployment and enable effective interaction with external environments; and (4) evaluation methods, encompassing benchmarks for key ability dimensions such as core language capabilities, reasoning, and safety, which support comprehensive and reliable assessment of model performance. Additionally, we identify critical research issues, including those concerning theoretical foundations, efficient scaling, alignment, and agentic capability, and highlight the open challenges they present. By synthesizing state-of-the-art insights and emerging trends, this survey aims to provide a systematic and comprehensive framework for understanding the trajectory, current limitations, and future directions of LLM progress.

Nowadays, navigation and ride-sharing apps have collected numerous images with spatio-temporal data. A core technology for analyzing such images, associated with spatio-temporal information, is Traffic Scene Understanding (TSU), which aims to provide a comprehensive description of the traffic scene. Unlike traditional spatio-temporal data analysis tasks, the dependence on both spatio-temporal and visual-textual data introduces distinct challenges to TSU task. However, recent research often treats TSU as a common image understanding task, ignoring the spatio-temporal information and overlooking the interrelations between different aspects of the traffic scene. To address these issues, we propose a novelSpatio-Temporal Enhanced Model based on CILP(ST-CLIP) for TSU. Our model uses the classic vision-language model, CLIP, as the backbone, and designs a Spatio-temporal Context Aware Multi-aspect Prompt (SCAMP) learning method to incorporate spatio-temporal information into TSU. The prompt learning method consists of two components: A dynamic spatio-temporal context representation module that extracts representation vectors of spatio-temporal data for each traffic scene image, and a bi-level ST-aware multi-aspect prompt learning module that integrates the ST-context representation vectors into word embeddings of prompts for the CLIP model. The second module also extracts low-level visual features and image-wise high-level semantic features to exploit interactive relations among different aspects of traffic scenes. To the best of our knowledge, this is the first attempt to integrate spatio-temporal information into vision-language models to facilitate TSU task. Experiments on two real-world datasets demonstrate superior performance in the complex scene understanding scenarios with a few-shot learning strategy.

Click-through rate (CTR) prediction holds a pivotal role in online advertising and recommendation systems, where even small improvements can significantly boost revenue. Existing research primarily focuses on designing dual-stream architectures to capture effective complex feature interactions from both explicit and implicit perspectives. However, these approaches are faced with two major challenges: (1) the high complexity of feature interaction learning, which increases computational demands and the overfitting risk, and (2) the imbalance between explicit and implicit modules, where one module’s output may dominate the final prediction. To address these issues, in this article, we propose dual-stream multi-layer perceptron (DS-MLP) , a novel feature interaction framework for the CTR prediction task. Specially, it leverages knowledge distillation to consolidate the capacity of learning explicit feature interaction into a main MLP network, while a parallel MLP simultaneously captures implicit feature interactions as a complement. To effectively optimize the DS-MLP architecture, we further design a specific learning approach with two alignment strategies for enhancing the compatibility of the two MLP components. Experiments demonstrate that DS-MLP , though merely a vanilla MLP structure (the final model), can achieve state-of-the-art performance across three widely used benchmarks, offering a scalable and efficient solution for large-scale recommendation systems. Our code is available at https://github.com/RUCAIBox/DS-MLP .

This is an erratum for the article “Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach” published in ACM Trans. Inf. Syst. 43, 5, Article 114 (July 2025), 37 pages.

Automatic prompt optimization is an important approach to improving the performance of large language models (LLMs). Recent research demonstrates the potential of using LLMs as prompt optimizers, which can generate improved task prompts via iterative refinement. In this paper, we propose a novel perspective to investigate the design of LLM-based prompt optimizers, by drawing an analogy with gradient-based model optimizers. To connect these two approaches, we identify two pivotal factors in model parameter learning: update direction and update method. By systematically analyzing a rich set of improvement strategies on the two aspects, we further develop a capable Gradient-inspired LLM-based Prompt Optimizer called GPO. At each step, it first retrieves relevant prompts from the optimization trajectory as the update direction. Then, it utilizes the generation-based refinement strategy to perform the update, while controlling the edit distance through a cosine-based decay strategy. Extensive experiments demonstrate the effectiveness and efficiency of GPO. In particular, GPO brings an additional improvement of up to 56.8% on Big-Bench Hard and 62.6% on MMLU compared to baseline methods.

Xinyu Tang, Xiaolei Wang, Zhihao Lv, Yingqian Min, Xin Zhao, Binbin Hu, Ziqi Liu, Zhiqiang Zhang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Continual pre-training (CPT) has been an important approach for adapting language models to specific domains or tasks. In this paper, we comprehensively study its key designs to balance the new abilities while retaining the original abilities, and present an effective CPT method that can greatly improve the Chinese language ability and scientific reasoning ability of LLMs. To achieve it, we design specific data mixture and curriculum strategies based on existing datasets and synthetic high-quality data. Concretely, we synthesize multidisciplinary scientific QA pairs based on related web pages to guarantee the data quality, and also devise the performance tracking and data mixture adjustment strategy to ensure the training stability. For the detailed designs, we conduct preliminary studies on a relatively small model, and summarize the findings to help optimize our CPT method. Extensive experiments on a number of evaluation benchmarks show that our approach can largely improve the performance of Llama-3 (8B), including both the general abilities (+8.81 on C-Eval and +6.31 on CMMLU) and the scientific reasoning abilities (+12.00 on MATH and +4.13 on SciEval). Our model, data, and codes are available at https://github.com/RUC-GSAI/Llama-3-SynE.

Due to the immense resource demands and the involved complex techniques, it is still challenging for successfully pre-training a large language models (LLMs) with state-of-the-art performance. In this paper, we explore the key bottlenecks and designs during pre-training, and make the following contributions: (1) a comprehensive investigation into the factors contributing to training instability; (2) a robust optimization approach designed to mitigate training instability effectively; (3) an elaborate data pipeline that integrates data synthesis, data curriculum, and data selection. By integrating the above techniques, we create a rather low-cost training recipe and use it to pre-train YuLan-Mini, a fully-open base model with 2.4B parameters on 1.08T tokens. Remarkably, YuLan-Mini achieves top-tier performance among models of similar parameter scale, with comparable performance to industry-leading models that require significantly more data. To facilitate reproduction, we release the full details of training recipe and data composition. Project details can be accessed at the following link: https://anonymous.4open.science/r/YuLan-Mini/README.md.

In this paper, we aim to improve the reasoning ability of large language models (LLMs) over knowledge graphs (KGs) to answer complex questions.Inspired by existing methods that design the interaction strategy between LLMs and KG, we propose an autonomous LLM-based agent framework, called KG-Agent, which enables a small LLM to actively make decisions until finishing the reasoning process over KGs.In KG-Agent, we integrate the LLM, multifunctional toolbox, KG-based executor, and knowledge memory, and develop an iteration mechanism that autonomously selects the tool and then updates the memory for reasoning over KG.To guarantee the effectiveness, we leverage program language to formulate the multi-hop reasoning process over the KG and synthesize a code-based instruction dataset to fine-tune the base LLM.Extensive experiments demonstrate that only using 10K samples for tuning LLaMA2-7B can outperform competitive methods using larger LLMs or more data, on both in-domain and out-domain datasets.Our code and data will be publicly released.

Recently, large language models (LLMs) have been introduced into recommender systems (RSs) as recommendation backbones or to enhance traditional recommendation models (TRMs). However, existing LLM-based RSs fail to fully leverage the complementary strengths of LLMs (e.g., world knowledge and reasoning capabilities) and TRMs (e.g., recommendation-specific knowledge and computational efficiency), resulting in shallow exploration of the item space. To address this limitation, we propose DeepRec, a novel LLM-based RS approach that facilitates autonomous multi-turn interactions between LLMs and TRMs for deep item space exploration. In each interaction turn, LLMs reason over user preferences and collaborate with TRMs to retrieve candidate items. After multi-turn interaction, LLMs rank the aggregated candidates to generate the final recommendations. We utilize reinforcement learning (RL) for optimization and introduce novel contributions in three key aspects: recommendation model based data rollout, recommendation-oriented hierarchical rewards, and a two-stage RL training strategy. For data rollout, we design a preference-aware TRM, with which LLMs interact to construct trajectory data. For reward design, we propose a hierarchical reward function that comprises both process-level and outcome-level rewards to optimize the interaction process and recommendation quality, respectively. For RL training, our two-stage RL strategy first guides LLMs to learn effective interactions with TRMs, followed by recommendation-oriented RL for performance enhancement. Experiments on public datasets show that DeepRec substantially outperforms both traditional and existing LLM-based baselines, establishing a new paradigm for deep exploration in recommender systems.

While modern recommender systems are instrumental in navigating information abundance, they remain fundamentally limited by static user modeling and reactive decision-making paradigms. Current large language model (LLM)-based agents inherit these shortcomings through their overreliance on heuristic pattern matching, yielding recommendations prone to shallow correlation bias, limited causal inference, and brittleness in sparse-data scenarios. We introduce STARec, a slow-thinking augmented agent framework that endows recommender systems with autonomous deliberative reasoning capabilities. Each user is modeled as an agent with parallel cognitions: fast response for immediate interactions and slow reasoning that performs chain-of-thought rationales. To cultivate intrinsic slow thinking, we develop anchored reinforcement training-a two-stage paradigm combining structured knowledge distillation from advanced reasoning models with preference-aligned reward shaping. This hybrid approach scaffolds agents in acquiring foundational capabilities (preference summarization, rationale generation) while enabling dynamic policy adaptation through simulated feedback loops. Experiments on MovieLens 1M and Amazon CDs benchmarks demonstrate that STARec achieves substantial performance gains compared with state-of-the-art baselines, despite using only 0.4% of the full training data.

Visual instruction tuning is crucial for enhancing the zero-shot generalization capability of Multi-modal Large Language Models (MLLMs). In this paper, we aim to investigate a fundamental question: “what makes for good visual instructions”. Through a comprehensive empirical study, we find that instructions focusing on complex visual reasoning tasks are particularly effective in improving the performance of MLLMs, with results correlating to instruction complexity. Based on this insight, we develop a systematic approach to automatically create high-quality complex visual reasoning instructions. Our approach employs a synthesize-complicate-reformulate paradigm, leveraging multiple stages to gradually increase the complexity of the instructions while guaranteeing quality. Based on this approach, we create the ComVint dataset with 32K examples, and fine-tune four MLLMs on it. Experimental results consistently demonstrate the enhanced performance of all compared MLLMs, such as a 27.86% and 27.60% improvement for LLaVA on MME-Perception and MME-Cognition, respectively. Our code and data are publicly available at the link: https://github.com/RUCAIBox/ComVint.

Large language models (LLMs) have shown impressive prowess in solving a wide range of tasks with world knowledge. However, it remains unclear how well LLMs are able to perceive their factual knowledge boundaries, particularly under retrieval augmentation settings. In this study, we present the first analysis on the factual knowledge boundaries of LLMs and how retrieval augmentation affects LLMs on open-domain question answering (QA), with a bunch of important findings. Specifically, we focus on three research questions and analyze them by examining QA, priori judgement and posteriori judgement capabilities of LLMs. We show evidence that LLMs possess unwavering confidence in their knowledge and cannot handle the conflict between internal and external knowledge well. Furthermore, retrieval augmentation proves to be an effective approach in enhancing LLMs’ awareness of knowledge boundaries. We further conduct thorough experiments to examine how different factors affect LLMs and propose a simple method to dynamically utilize supporting documents with our judgement strategy. Additionally, we find that the relevance between the supporting documents and the questions significantly impacts LLMs’ QA and judgemental capabilities.

Large reasoning models (LRMs) have exhibited strong performance on complex reasoning tasks, with further gains achievable through increased computational budgets at inference.However, current test-time scaling methods predominantly rely on redundant sampling, ignoring the historical experience utilization, thereby limiting computational efficiency.To overcome this limitation, we propose Sticker-TTS, a novel test-time scaling framework that coordinates three collaborative LRMs to iteratively explore and refine solutions guided by historical attempts.At the core of our framework are distilled key conditions-termed stickers-which drive the extraction, refinement, and reuse of critical information across multiple rounds of reasoning.To further enhance the efficiency and performance of our framework, we introduce a two-stage optimization strategy that combines imitation learning with self-improvement, enabling progressive refinement.Extensive evaluations on three challenging mathematical reasoning benchmarks, including AIME-24, AIME-25, and Olym-MATH, demonstrate that Sticker-TTS consistently surpasses strong baselines, including self-consistency and advanced reinforcement learning approaches, under comparable inference budgets.These results highlight the effectiveness of sticker-guided historical experience utilization.

Multi-lingual ability transfer has become increasingly important for the broad application of large language models (LLMs).Existing work highly relies on training with the multi-lingual ability-related data, which may not be available for low-resource languages.To solve it, we propose a Multilingual Abilities Extraction and Combination approach (MAEC), which decomposes and extracts language-agnostic ability-related weights from LLMs, and combines them across different languages by simple addition and subtraction operations without training.Specifically, our MAEC consists of the extraction and combination stages.In the extraction stage, we firstly locate key neurons that are highly related to specific abilities, and then employ them to extract the transferable ability-related weights.In the combination stage, we further select the ability-related tensors that mitigate the linguistic effects, and design a combining strategy based on them and the languagespecific weights, to build the multi-lingual ability-enhanced LLM.To assess the effectiveness of our approach, we conduct extensive experiments on LLaMA-3 8B on mathematical and scientific tasks in both high-resource and low-resource lingual scenarios.Empirical results have shown that MAEC can effectively and efficiently extract and combine the advanced abilities, achieving comparable performance with PaLM.Resources are available at https://github.com/RUCAIBox/MAET.

Advancements in Large Language Models (LLMs) have extended their input context length, yet they still struggle with retrieval and reasoning in long-context inputs.Existing methods propose to utilize the prompt strategy and Retrieval-Augmented Generation (RAG) to alleviate this limitation.However, they still face challenges in balancing retrieval precision and recall, impacting their efficacy in answering questions.To address this, we introduce CAFE, a two-stage coarse-to-fine method to enhance multi-document questionanswering capacities.By gradually eliminating the negative impacts of background and distracting documents, CAFE makes the responses more reliant on the evidence documents.Initially, a coarse-grained filtering method leverages retrieval heads to identify and rank relevant documents.Then, a fine-grained steering method guides attention to the most relevant content.Experiments across benchmarks show that CAFE outperforms baselines, achieving an average SubEM improvement of up to 22.1% and 13.7% over SFT and RAG methods, respectively, across three different models.Our code is available at https://github. com/RUCAIBox/CAFE.

Despite the impressive chain-of-thought (CoT) reasoning ability of large language models (LLMs), its underlying mechanisms remains unclear.In this paper, we explore the inner workings of LLM's CoT ability via the lens of neurons in the feed-forward layers.We propose an efficient method to identify reasoningcritical neurons by analyzing their activation patterns under reasoning chains of varying quality.Based on it, we devise a rather simple intervention method that directly stimulates these reasoning-critical neurons, to guide the generation of high-quality reasoning chains.Extended experiments validate the effectiveness of our method and demonstrate the critical role these identified neurons play in CoT reasoning.

The goal of transferable sequential recommendation (TSR) is to improve the performance of sequential recommenders in multiple target domains leveraging knowledge transferred from source domains. Most existing transferable sequential recommenders rely on item modality information but pay insufficient attention to the significant differences across domains (i.e., distinctive user behavior characteristics) in recommender systems. Besides, they are typically built on specific pre-trained model structures, limiting their model generalization to various different downstream domains and sequential recommenders in practical applications. This paper aims to address these issues by exploring the potential of mixture-of-LoRAs based domain adaption and an ensemble of sequential recommenders for learning generalizable and transferable domain-adaptive item representations. We propose MoL-TSR, a pre-trained plug-in mixture of low-rank adaption model for TSR. Specifically, we first devise a mixture-ofLoRAs based domain projector to produce domain-adaptive item representations based on both item modality and user behavior information for recommendation, enhancing the item discrimination and domain transfer ability. Then, we pre-train an ensemble of sequential recommendation models with contrastive learning to enhance the model-aspect generalization. As a general pretrained model, the mixture-of-LoRAs based domain projector in our MoL-TSR is a plug-in item semantic representation learning module, which is convenient to migrate into other transferable sequential recommenders to enhance item semantics better suited for recommendation. Extensive experiments demonstrate the effectiveness and model generalization of MoL-TSR, promising a practical solution for learning generalizable and transferable item semantic representations in recommendation. Our code is available at https://github.com/RUCAIBox/MoL-TSR.

Video Multimodal Large Language Models~(MLLMs) have shown remarkable capability of understanding the video semantics on various downstream tasks. Despite the advancements, there is still a lack of systematic research on visual context representation, which refers to the scheme to select frames from a video and further select the tokens from a frame. In this paper, we explore the design space for visual context representation, and aim to improve the performance of video MLLMs by finding more effective representation schemes. Firstly, we formulate the task of visual context representation as a constrained optimization problem, and model the language modeling loss as a function of the number of frames and the number of embeddings (or tokens) per frame, given the maximum visual context window size. Then, we explore the scaling effects in frame selection and token selection respectively, and fit the corresponding function curve by conducting extensive empirical experiments. We examine the effectiveness of typical selection strategies and present empirical findings to determine the two factors. Furthermore, we study the joint effect of frame selection and token selection, and derive the optimal formula for determining the two factors. We demonstrate that the derived optimal settings show alignment with the best-performed results of empirical experiments. The data and code are available at: https://github.com/RUCAIBox/Opt-Visor.

The emergence of in-context learning (ICL) is potentially attributed to two major abilities: task recognition (TR) for recognizing the task from demonstrations and utilizing pre-trained priors, and task learning (TL) for learning from demonstrations. However, relationships between the two abilities and how such relationships affect the emergence of ICL is unclear. In this paper, we take the first step by examining the pre-training dynamics of the emergence of ICL. With carefully designed metrics, we find that these two abilities are, in fact, competitive during pre-training. Moreover, we observe a negative correlation between the competition and the performance of ICL. Further analysis of common pre-training factors (i.e., model size, dataset size, and data curriculum) demonstrates possible ways to regulate the competition. Based on these insights, we propose a simple yet effective method to better integrate these two abilities for ICL at inference time. Through adaptive ensemble learning, the performance of ICL can be significantly boosted, enabling two small models to outperform a larger one with more than twice the parameters.

Large language models (LLMs) have become increasingly proficient at simulating various personality traits, an important capability for supporting related applications (e.g., role-playing). To further improve this capacity, in this paper, we present a neuron based approach for personality trait induction in LLMs, with three major technical contributions. First, we construct PERSONALITYBENCH, a large-scale dataset for identifying and evaluating personality traits in LLMs. This dataset is grounded in the Big Five personality traits from psychology and designed to assess the generative capabilities of LLMs towards specific personality traits. Second, by leveraging PERSONALITYBENCH, we propose an efficient method for identifying personality-related neurons within LLMs by examining the opposite aspects of a given trait. Third, we develop a simple yet effective induction method that manipulates the values of these identified personality-related neurons, which enables fine-grained control over the traits exhibited by LLMs without training and modifying model parameters. Extensive experiments validates the efficacy of our neuron identification and trait induction methods. Notably, our approach achieves comparable performance as fine-tuned models, offering a more efficient and flexible solution for personality trait induction in LLMs.

Adapting large language models (LLMs) to specialized domains typically requires domain-specific corpora for continual pre-training to facilitate knowledge memorization and related instructions for fine-tuning to apply this knowledge. However, this method may lead to inefficient knowledge memorization due to a lack of awareness of knowledge utilization during the continual pre-training and demands LLMs to simultaneously learn knowledge utilization and format alignment with divergent training objectives during the fine-tuning. To enhance the domain adaptation of LLMs, we revise this process and propose a new domain adaptation framework including domain knowledge learning and general format alignment, called \emph{Mix-CPT}. Specifically, we first conduct a knowledge mixture continual pre-training that concurrently focuses on knowledge memorization and utilization. To avoid catastrophic forgetting, we further propose a logit swap self-distillation constraint. By leveraging the knowledge and capabilities acquired during continual pre-training, we then efficiently perform instruction tuning and alignment with a few general training samples to achieve format alignment. Extensive experiments show that our proposed \emph{Mix-CPT} framework can simultaneously improve the task-solving capabilities of LLMs on the target and general domains.

Retrieval-augmented generation (RAG) systems face a fundamental challenge in aligning independently developed retrievers and large language models (LLMs). Existing approaches typically involve modifying either component or introducing simple intermediate modules, resulting in practical limitations and sub-optimal performance. Inspired by human search behavior—typically involving a back-and-forth process of proposing search queries and reviewing documents, we propose C-3PO, a proxy-centric framework that facilitates communication between retrievers and LLMs through a lightweight multi-agent system. Our framework implements three specialized agents that collaboratively optimize the entire RAG pipeline without altering the retriever and LLMs. These agents work together to assess the need for retrieval, generate effective queries, and select information suitable for the LLMs. To enable effective multi-agent coordination, we develop a tree-structured rollout approach for reward credit assignment in reinforcement learning. Extensive experiments in both in-domain and out-of-distribution scenarios demonstrate that C-3PO significantly enhances RAG performance while maintaining plug-and-play flexibility and superior generalization capabilities.

In recent years, substantial research efforts have been devoted to enhancing sequential recommender systems by integrating abundant side information with ID-based collaborative information. This study specifically focuses on leveraging the textual metadata (e.g., titles and brands) associated with items. While existing methods have achieved notable success by combining text and ID representations, they often struggle to strike a balance between textual information embedded in text representations and collaborative information from sequential patterns of user behavior. In light of this, we propose CCFRec, a novel Code-based textual and Collaborative semantic Fusion method for sequential Recommendation. The key idea behind our approach is to bridge the gap between textual and collaborative information using semantic codes. Specifically, we generate fine-grained semantic codes from multi-view text embeddings through vector quantization techniques. Subsequently, we develop a code-guided semantic-fusion module based on the cross-attention mechanism to flexibly extract and integrate relevant information from text representations. In order to further enhance the fusion of textual and collaborative semantics, we introduce an optimization strategy that employs code masking with two specific objectives: masked code modeling and masked sequence alignment. The merit of these objectives lies in leveraging mask prediction tasks and augmented item representations to capture code correlations within individual items and enhance the sequence modeling of the recommendation backbone. Extensive experiments conducted on four public datasets demonstrate the superiority of CCFRec, showing significant improvements over various sequential recommendation models. Our code is available at https://github.com/RUCAIBox/CCFRec.

Visual instruction tuning is the key to building large vision language models (LVLMs), which can greatly improve the task solving and generalization capabilities. Previous work mostly collects a mixture of existing visual instruction datasets via heuristic ways for train- ing (even more than a million instructions), which may introduce data redundancy and increase the training cost. To investigate it, we conduct a series of empirical studies, which show that greatly reducing the amount of instructions from several tasks even do not affect the performance, indicating significant redundancy within the visual instruction datasets. Based on the findings, we propose a high-value data selection approach TIVE, to eliminate redundancy within the visual instruction data and reduce the training cost. In TIVE, based on the gradient-based influence functions, we estimate the instance influence score on its corresponding task and the task difficulty score. Then, we leverage these scores to determine the task proportion within the visual instruction subset, and select high-value instances for each task, respectively. Experiments on various LVLMs show that our approach using only about 15% data can achieve comparable performance to the full-data fine-tuned model across eight benchmarks, even surpassing it on four of the benchmarks. Our code and data will be publicly released.

Jinhao Jiang, Jiayi Chen, Junyi Li, Ruiyang Ren, Shijie Wang, Xin Zhao, Yang Song, Tao Zhang. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Xinyu Tang, Xiaolei Wang, Xin Zhao, Ji-Rong Wen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Large reasoning models (LRMs) have demonstrated strong performance on complex reasoning tasks, but often suffer from overthinking, generating redundant content regardless of task difficulty. Inspired by the dual process theory in cognitive science, we propose Adaptive Cognition Policy Optimization (ACPO), a reinforcement learning framework that enables LRMs to achieve efficient reasoning through adaptive cognitive allocation and dynamic system switch. ACPO incorporates two key components: (1) introducing system-aware reasoning tokens to explicitly represent the thinking modes thereby making the model's cognitive process transparent, and (2) integrating online difficulty estimation and token length budget to guide adaptive system switch and reasoning during reinforcement learning. To this end, we propose a two-stage training strategy. The first stage begins with supervised fine-tuning to cold start the model, enabling it to generate reasoning paths with explicit thinking modes. In the second stage, we apply ACPO to further enhance adaptive system switch for difficulty-aware reasoning. Experimental results demonstrate that ACPO effectively reduces redundant reasoning while adaptively adjusting cognitive allocation based on task complexity, achieving efficient hybrid reasoning.

Mixture-of-Experts (MoE) models achieve a favorable trade-off between performance and inference efficiency by activating only a subset of experts. However, the memory overhead of storing all experts remains a major limitation, especially in large-scale MoE models such as DeepSeek-R1 (671B). In this study, we investigate domain specialization and expert redundancy in large-scale MoE models and uncover a consistent behavior we term~\emph{few-shot expert localization}, with only a few in-domain demonstrations, the model consistently activates a sparse and stable subset of experts on tasks within the same domain. Building on this observation, we propose a simple yet effective pruning framework, \textbf{EASY-EP}, that leverages a few domain-specific demonstrations to identify and retain only the most relevant experts. EASY-EP comprises two key components: \textbf{output-aware expert importance assessment} and \textbf{expert-level token contribution estimation}. The former evaluates the importance of each expert for the current token by considering the gating scores and L2 norm of the outputs of activated experts, while the latter assesses the contribution of tokens based on representation similarities before and after routed experts. Experiments on DeepSeek-R1 and DeepSeek-V3-0324 show that our method can achieve comparable performances and $2.99\times$ throughput under the same memory budget as the full model, with only half the experts. Our code is available at https://github.com/RUCAIBox/EASYEP.

In this paper, we propose a novel optimizer state compression algorithm, namely \textbf{$\pi$-Quant}, which leverages the properties of irrational numbers (\eg $\pi$) for memory-efficient training. The core idea is based on our mathematical findings, which show that a pair of parameters can be represented by a single rotation angle using the complex rotation scheme. Building on this insight, we map the parameters into a complex space and perform quantization using the corresponding rotation angles. To efficiently integrate it into optimization process, we develop an efficient system of geometric equations that computes the precise rotation angles with linear complexity. We evaluate $\pi$-Quant on a wide range of tasks. Our experiments show that it can reduce the bit-width of parameters to 3.32-bit, achieving a 41.8\% decrease in GPU memory usage, all while maintaining full accuracy. \textcolor{blue}{We have submitted the code in supplementary materials}.

With the significant progress of large reasoning models in complex coding and reasoning tasks, existing benchmarks, like LiveCodeBench and CodeElo, are insufficient to evaluate the coding capabilities of large language models (LLMs) in real competition environments. Moreover, current evaluation metrics such as Pass@K fail to capture the reflective abilities of reasoning models. To address these challenges, we propose ICPC-Eval, a top-level competitive coding benchmark designed to probing the frontiers of LLM reasoning. ICPC-Eval includes 118 carefully curated problems from 11 recent ICPC contests held in various regions of the world, offering three key contributions: 1) A challenging realistic ICPC competition scenario, featuring a problem type and difficulty distribution consistent with actual contests. 2) A robust test case generation method and a corresponding local evaluation toolkit, enabling efficient and accurate local evaluation. 3) An effective test-time scaling evaluation metric, Refine@K, which allows iterative repair of solutions based on execution feedback. The results underscore the significant challenge in evaluating complex reasoning abilities: top-tier reasoning models like DeepSeek-R1 often rely on multi-turn code feedback to fully unlock their in-context reasoning potential when compared to non-reasoning counterparts. Furthermore, despite recent advancements in code generation, these models still lag behind top-performing human teams. We release the benchmark at: https://github.com/RUCAIBox/ICPC-Eval

Generative recommender systems have gained increasing attention as an innovative approach that directly generates item identifiers for recommendation tasks. Despite their potential, a major challenge is the effective construction of item identifiers that align well with recommender systems. Current approaches often treat item tokenization and generative recommendation training as separate processes, which can lead to suboptimal performance. To overcome this issue, we introduce ETEGRec, a novel End-To-End Generative Recommender that unifies item tokenization and generative recommendation into a cohesive framework. Built on a dual encoder-decoder architecture, ETEGRec consists of an item tokenizer and a generative recommender. To enable synergistic interaction between these components, we propose a recommendation-oriented alignment strategy, which includes two key optimization objectives: sequence-item alignment and preference-semantic alignment. These objectives tightly couple the learning processes of the item tokenizer and the generative recommender, fostering mutual enhancement. Additionally, we develop an alternating optimization technique to ensure stable and efficient end-to-end training of the entire framework. Extensive experiments demonstrate the superior performance of our approach compared to traditional sequential recommendation models and existing generative recommendation baselines. Our code is available at https://github.com/RUCAIBox/ETEGRec.

Conversational recommendation systems (CRSs) use multi-turn interaction to capture user preferences and provide personalized recommendations. A fundamental challenge in CRSs lies in effectively understanding user preferences from conversations. Previous research primarily focuses on the issue of insufficient contextual information in conversations. They address this by introducing external knowledge sources, such as knowledge graphs, large language models (LLMs), and conversational recommendation corpora. Based on this, they design specific alignment strategies (e.g., prompt learning and instruction tuning) to integrate such knowledge for user preference understanding and item recommendation. However, user preferences can be multifaceted and complex, posing significant challenges for accurate recommendations even with access to abundant external knowledge. While interaction with users can clarify their true preferences, frequent user involvement may lead to a degraded user experience.

Considering the inherent limitations of parametric knowledge in large language models (LLMs), retrieval-augmented generation (RAG) is widely employed to expand their knowledge scope. Since RAG has shown promise in knowledge-intensive tasks like open-domain question answering, its broader application to complex tasks and intelligent assistants has further advanced its utility. Despite this progress, the underlying knowledge utilization mechanisms of LLM-based RAG remain underexplored. In this paper, we present a systematic investigation of the intrinsic mechanisms by which LLMs integrate internal (parametric) and external (retrieved) knowledge in RAG scenarios. Specially, we employ knowledge stream analysis at the macroscopic level, and investigate the function of individual modules at the microscopic level. Drawing on knowledge streaming analyses, we decompose the knowledge utilization process into four distinct stages within LLM layers: knowledge refinement, knowledge elicitation, knowledge expression, and knowledge contestation. We further demonstrate that the relevance of passages guides the streaming of knowledge through these stages. At the module level, we introduce a new method, knowledge activation probability entropy (KAPE) for neuron identification associated with either internal or external knowledge. By selectively deactivating these neurons, we achieve targeted shifts in the LLM's reliance on one knowledge source over the other. Moreover, we discern complementary roles for multi-head attention and multi-layer perceptron layers during knowledge formation. These insights offer a foundation for improving interpretability and reliability in retrieval-augmented LLMs, paving the way for more robust and transparent generative solutions in knowledge-intensive domains.

Personalized recommender systems, which focus on predicting users' interests, have significantly enhanced user experiences across diverse applications. However, existing approaches implicitly model users' preferences through fitting the fine-grained labels (e.g., click labels), but often neglecting the coarse-grained interest information inherent in the inputs themselves. Relying solely on the fine-grained labels could bring negative impact on interest modeling and limit the performance, as the labels may carry inevitable noise in real-world scenarios. In addition, it is considerably demanding in terms of data for most existing approaches to effectively model users' multi-granularity interests with limited or no supporting examples, resulting in subpar performance due to the significant long-tail phenomenon. To tackle these issues, we propose a novel learning framework named the Multi-Granularity Interest Prediction Framework (MGIPF), for better modeling users' diverse interests. Unlike prior work, our key idea is to utilize both the coarse-grained and fine-grained interests for supervising the training of models. Specifically, we introduce a pseudo-labeling approach explicitly mining users' potential multi-granularity interests from the raw data, and propose coarse-grained interest prediction modules that collaborate to utilize the multi-granularity supervision signals to enhance the learning of low-frequency items. The corresponding coarse-grained losses are softly weighted, taking into account the varying confidence of potential multi-granularity preferences on positive and negative samples. Importantly, our framework is lightweight and adaptable, capable of being applied effectively to mainstream recommendation models, establishing a comprehensive end-to-end training process. Extensive experiments conducted on three publicly available datasets have demonstrated the efficacy of our approach. The code is available at https://github.com/GeWu-Lab/MGIPF.

In the era of vast digital information, the sheer volume and heterogeneity of available information present significant challenges for intricate information seeking. Users frequently face multistep web search tasks that involve navigating vast and varied data sources. This complexity demands every step remains comprehensive, accurate, and relevant. However, traditional search methods often struggle to balance the need for localized precision with the broader context required for holistic understanding, leaving critical facets of intricate queries underexplored. In this paper, we introduce an LLM-based search assistant that adopts a new information seeking paradigm with holistically guided Monte Carlo tree search (HG-MCTS). We reformulate the task as a progressive information collection process with a knowledge memory and unite an adaptive checklist with multi-perspective reward modeling in MCTS. The adaptive checklist provides explicit sub-goals to guide the MCTS process toward comprehensive coverage of complex user queries. Simultaneously, our multi-perspective reward modeling offers both exploration and retrieval rewards, along with progress feedback that tracks completed and remaining sub-goals, refining the checklist as the tree search progresses. By striking a balance between localized tree expansion and global guidance, HG-MCTS reduces redundancy in search paths and ensures that all crucial aspects of an intricate query are properly addressed. Extensive experiments on real-world intricate information seeking tasks demonstrate that HG-MCTS acquires thorough knowledge collections and delivers more accurate final responses compared with existing baselines.

Large language models (LLMs), with advanced linguistic capabilities, have been employed in reranking tasks through a sequence-to-sequence approach. In this paradigm, multiple passages are reranked in a listwise manner and a textual reranked permutation is generated. However, due to the limited context window of LLMs, this reranking paradigm requires a sliding window strategy to iteratively handle larger candidate sets. This not only increases computational costs but also restricts the LLM from fully capturing all the comparison information for all candidates. To address these challenges, we propose a novel self-calibrated listwise reranking method, which aims to leverage LLMs to produce global relevance scores for ranking. To achieve it, we first propose the relevance-aware listwise reranking framework, which incorporates explicit list-view relevance scores to improve reranking efficiency and enable global comparison across the entire candidate set. Second, to ensure the comparability of the computed scores, we propose self-calibrated training that uses point-view relevance assessments generated internally by the LLM itself to calibrate the list-view relevance assessments. Extensive experiments and comprehensive analysis on the BEIR benchmark and TREC Deep Learning Tracks demonstrate the effectiveness and efficiency of our proposed method.

Recently, many efforts have been devoted to building effective sequential recommenders.Despite their effectiveness, these methods typically develop a single model to serve all users.However, our empirical studies reveal that different sequential encoders have intrinsic architectural biases and tend to focus on specific behavioral patterns, i.e., particular frequency range of user behavior sequences.For example, the Self-Attention module is essentially a low-pass filter, focusing on low-frequency information while neglecting the high-frequency details.This evidently limits their ability to capture diverse user patterns, leading to suboptimal recommendations.To tackle this problem, we present FamouSRec, a Frequency-Augmented Mixture-of-Heterogeneous-Experts Framework for personalized Recommendations.Our approach builds an MoEbased recommender system, integrating the strengths of various experts to achieve diversified user modeling.For developing the MoE framework, as the key to our approach, we instantiate experts with various model architectures, aiming to leverage their inherent architectural biases and capture diverse behavioral patterns.For selecting appropriate experts to serve individuals, we introduce a frequency-augmented router.It first identifies frequency components in user behavior sequences that are suited for expert encoding, and then conducts customized routing based on the informativeness of these components.Building on this framework, we further propose two novel contrastive tasks to enhance expert

Recently, Large Language Models (LLMs) such as ChatGPT have showcased remarkable abilities in solving general tasks, demonstrating the potential for applications in recommender systems. To assess how effectively LLMs can be used in recommendation tasks, our study primarily focuses on employing LLMs as recommender systems through prompt engineering. We propose a general framework for leveraging LLMs in recommendation tasks, focusing on the capabilities of LLMs as recommenders. To conduct our analysis, we formalize the input of LLMs for recommendation into natural language prompts with two key aspects and explain how our framework can be generalized to various recommendation scenarios. As for the use of LLMs as recommenders, we analyze the impact of public availability, tuning strategies, model architecture, parameter scale, and context length on recommendation results based on the classification of LLMs. As for prompt engineering, we further analyze the impact of four important components of prompts, i.e., task descriptions, user interest modeling, candidate items construction, and prompting strategies. In each section, we first define and categorize concepts in line with the existing literature. Then, we propose inspiring research questions followed by detailed experiments on two public datasets, in order to systematically analyze the impact of different factors on recommendation performance. Based on our empirical analysis, we finally summarize promising directions to shed lights on future research.

Cross-domain sequential recommendation (CDSR) utilizes data from multiple domains to recommend the user’s next interaction based on his latest interaction sequence. Currently, many cross-domain sequential recommendation algorithms have been proven to achieve good recommendation performance. However, these algorithms overlook the influence of users’ long-term behavioral patterns and general interests when extracting their current preferences. In this article, we propose a Hierarchical Gating Network for Cross-Domain Sequential Recommendation (HGNCDSR). Specifically, we simultaneously train single-domain and cross-domain interaction sequences, utilizing a hierarchical gating network to capture user interest representations in single-domain and cross-domain, respectively. A feature gating and an instance gating are applied respectively to extract user interests at item feature level and instance level. While learning current preferences from behavior sequences, user representations that reflect behavioral patterns and general interests are simultaneously learned and strengthened. Additionally, we employ the item–item product to model the relationships between candidate items and those in the interaction sequence. Both current interests and item relevance are considered simultaneously, integrating single-domain and cross-domain user preferences to predict the user’s next interaction. We design extensive experiments to show that HGNCDSR has better recommendation performance than other state-of-the-art models.

Simulating high quality user behavior data has always been a fundamental yet challenging problem in human-centered applications such as recommendation systems, social networks, among many others. The major difficulty of user behavior simulation originates from the intricate mechanism of human cognitive and decision processes. Recently, substantial evidence has suggested that by learning huge amounts of web knowledge, large language models (LLMs) can achieve human-like intelligence and generalization capabilities. Inspired by such capabilities, in this article, we take an initial step to study the potential of using LLMs for user behavior simulation in the recommendation domain. To make LLMs act like humans, we design profile, memory and action modules to equip them, building LLM-based agents to simulate real users. To enable interactions between different agents and observe their behavior patterns, we design a sandbox environment, where each agent can interact with the recommendation system, and different agents can converse with their friends via one-to-one chatting or one-to-many social broadcasting. In the experiments, we first demonstrate the believability of the agent-generated behaviors based on both subjective and objective evaluations. Then, to show the potential applications of our method, we simulate and study two social phenomena including (1) information cocoons and (2) user conformity behaviors. We find that controlling the personalization degree of recommendation algorithms and improving the heterogeneity of user social relations can be two effective strategies for alleviating the problem of information cocoon, and the conformity behaviors can be highly influenced by the amount of user social relations. To advance this direction, we have released our project at https://github.com/RUC-GSAI/YuLan-Rec .

With the widespread deployment of recommender systems on various online platforms, researchers are striving to develop transferable recommendation algorithms that can effectively adapt to new task scenarios without requiring the re-training of new recommenders. However, there have been challenges in dealing with explicit ID modeling in this context. Recently, researchers have drawn inspiration from the achievements of pre-trained language models (PLMs), making it possible to acquire ID-agnostic representations by utilizing the corresponding texts of items. These representations have shown to be transferable across diverse domains. However, while these methods demonstrate generalization, they are less proficient in making personalized recommendations as they learn universal representation. In light of this issue, we present a review-enhanced universal sequence representation learning approach named RUNSRec . Our goal is to not only comprehend universal user behavioral patterns across different domains but also capture their inherent preferences to make recommendations. Our approach makes three technical advancements toward this objective. Firstly, we introduce a lightweight item encoding architecture based on parametric whitening and mixture-of-experts enhanced adapter. It learns discriminative item textual representations by encoding their corresponding identity text and review text, with a discriminative keyword extraction method to enhance the representation identifiability. Secondly, we propose a universal sequence representation learning method that enables the training of transferable recommenders across diverse domains, based on two novel contrastive learning tasks. Furthermore, we introduce a personalized adapter tuning mechanism that enables the universal recommender to capture user personal preferences in a parameter-efficient way. By incorporating universal behavioral patterns learned during the pre-training stage and personalized user tastes captured through adapter tuning, our approach achieves a better balance between generalization and personalization in transferable recommender systems. Extensive experiments conducted on five real-world datasets have demonstrated the effectiveness of our proposed approach.

In the past few decades, recommender systems have attracted much attention in both research and industry communities. Existing recommendation models mainly learn the underlying user preference from historical behavior data (typically in the forms of item IDs), and then estimate the user–item matching relationships for recommendations. Inspired by the recent progress on large language models (LLMs), we develop a different recommendation paradigm, considering recommendation as instruction following by LLMs. The key idea is that the needs of a user can be expressed in natural language descriptions (called instructions ), so that LLMs can understand and further execute the instruction for fulfilling the recommendation. For this purpose, we instruction tune the 3B Flan-T5-XL, to better adapt LLMs to recommender systems. We first design a general instruction format for describing the preference, intention, and task form of a user in natural language. Then we manually design 39 instruction templates and automatically generate large amounts of user-personalized instruction data with varying types of preferences and intentions. To demonstrate the effectiveness of our approach, we instantiate the instructions into several widely studied recommendation (or search) tasks, and conduct extensive experiments with real-world datasets. Experiment results show that our approach can outperform several competitive baselines, including the powerful GPT-3.5, on these evaluation tasks. Our approach sheds light on developing user-friendly recommender systems, in which users can freely communicate with the system and obtain accurate recommendations via natural language instructions.

Junyi Li, Jie Chen, Ruiyang Ren, Xiaoxue Cheng, Xin Zhao, Jian-Yun Nie, Ji-Rong Wen. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Key-value (KV) caching is an important technique to accelerate the inference of large language models (LLMs), but incurs significant memory overhead.To compress the size of KV cache, existing methods often compromise precision or require extra data for calibration, limiting their practicality in LLM deployment.In this paper, we introduce DecoQuant, a novel data-free low-bit quantization technique based on tensor decomposition methods, to effectively compress KV cache.Our core idea is to adjust the outlier distribution of the original matrix by performing tensor decomposition, so that the quantization difficulties are migrated from the matrix to decomposed local tensors.Specially, we find that outliers mainly concentrate on small local tensors, while large tensors tend to have a narrower value range.Based on this finding, we propose to apply low-bit quantization to the large tensor, while maintaining high-precision representation for the small tensor.Furthermore, we utilize the proposed quantization method to compress the KV cache of LLMs to accelerate the inference and develop an efficient dequantization kernel tailored specifically for DecoQuant.Through extensive experiments, DecoQuant demonstrates remarkable efficiency gains, showcasing up to a ∼75% reduction in memory footprint while maintaining comparable generation quality.

Yuchong Sun, Che Liu, Kun Zhou, Jinwen Huang, Ruihua Song, Xin Zhao, Fuzheng Zhang, Di Zhang, Kun Gai. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Tianyi Tang, Wenyang Luo, Haoyang Huang, Dongdong Zhang, Xiaolei Wang, Xin Zhao, Furu Wei, Ji-Rong Wen. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Chain-of-Thought (CoT) prompting can enhance the reasoning capabilities of large language models (LLMs), establishing itself as a primary approach to solving complex reasoning tasks. Existing CoT synthesis approaches usually focus on simpler reasoning tasks and thus result in low-quality and inconsistent CoT prompts. In response to this challenge, we present an empirical investigation of CoT prompting and introduce CoTGenius, a novel framework designed for the automatic generation of superior CoT prompts. CoTGenius is developed based on three major evolution strategies, i.e., complicate, diversify, and specify—alongside two filtering mechanisms: evolutionary success judgement and correctness verification. We further employ CoTGenius to create an extensive CoT dataset, and subsequently fine-tune the Llama 2-Chat 7B and 13B models on this dataset. We call the resulting model ChainLM. To deal with the cumulative error issue in reasoning steps, we propose a step-level debating method, wherein multiple debaters discuss each reasoning step to arrive at the correct answer. Extensive experiments demonstrate that our ChainLM models exhibit enhanced proficiency in addressing a spectrum of complex reasoning problems compared to existing models. In addition, we conduct an in-depth analysis of the impact of data categories within CoTGenius on the model performance. We release our dataset and code at https://github.com/RUCAIBox/ChainLM.

Large language models (LLMs) have achieved dramatic proficiency over NLP tasks with normal length. Recently, multiple studies have committed to extending the context length and enhancing the long text modeling capabilities of LLMs. To comprehensively evaluate the long context ability of LLMs, we propose BAMBOO, a multi-task long context benchmark. BAMBOO has been designed with four principles: comprehensive capacity evaluation, avoidance of data contamination, accurate automatic evaluation, and different length levels. It consists of 10 datasets from 5 different long text understanding tasks, i.e., question answering, hallucination detection, text sorting, language modeling, and code completion, to cover various domains and core capacities of LLMs. We conduct experiments with five widely-used long-context models and further discuss five key questions for long text research. In the end, we discuss problems of current long-context models and point out future directions for enhancing long text modeling capacities. We release our data, prompts, and code at https://anonymous.4open.science/r/BAMBOO/.

Lightweight fine-tuning is widely used as an important technique for efficiently adapting pre-trained language models (PLM) to downstream tasks. Despite the reduction in trainable parameters, existing lightweight fine-tuning methods are found to be effective in low-resource settings but often fail in high-resource settings, leading to unreliable outcomes. This limitation can be attributed to inflexible strategies: they identify the parameters of the model to be trained before fine-tuning and remain unchanged without taking into account the inherent variance of generalization ability in model components (i.e., feed-forward, attention layers) and potential changes during the fine-tuning process. In this paper, we introduce a simple but effective calibration for lightweight fine-tuning PLMs based on the matrix’s stable rank according to both model components and the training process. We proposed both theoretical analyses and experimental verification for the proposed calibration strategy. Considering efficiency, we further propose time-aware and structure-aware strategies to determine the most crucial time to commence the fine-tuning procedure and selectively apply parameter matrices for lightweight fine-tuning, respectively. Extensive experiments demonstrate the superiority of our proposed fine-tuning approach (average improvement 3.1 for GLUE score compared to lightweight fine-tuning method).

Despite the superior performance, Large Language Models (LLMs) require significant computational resources for deployment and use. To overcome this issue, quantization methods have been widely applied to reduce the memory footprint of LLMs as well as increase the inference rate. However, a major challenge is that low-bit quantization methods often lead to performance degradation. It is important to understand how quantization impacts the capacity of LLMs. Different from previous studies focused on overall performance, this work aims to investigate the impact of quantization on emergent abilities, which are important characteristics that distinguish LLMs from small language models. Specifically, we examine the abilities of in-context learning, chain-of-thought reasoning, and instruction-following in quantized LLMs. Our empirical experiments show that these emergent abilities still exist in 4-bit quantization models, while 2-bit models encounter severe performance degradation on the test of these abilities. To improve the performance of low-bit models, we conduct two special experiments: (1) fine-gained impact analysis that studies which components (or substructures) are more sensitive to quantization, and (2) performance compensation through model fine-tuning. Our work derives a series of important findings to understand the impact of quantization on emergent abilities and sheds light on the possibilities of extremely low-bit quantization for LLMs.

Recently, continuous diffusion models (CDM) have been introduced into non-autoregressive (NAR) text-to-text generation.However, the discrete nature of text increases the difficulty of CDM to generate coherent and fluent texts, and also causes the incompatibility problem between CDM and advanced NLP techniques, especially the popular pre-trained language models (PLMs).To solve it, we propose Diffusion-NAT, which introduces discrete diffusion models (DDM) into NAR text-to-text generation and integrates BART to improve the performance.By revising the decoding process of BART and the typical settings of DDM, we unify the inference process of BART and the denoising process of DDM into the same NAR masked tokens recovering task.In this way, DDM can rely on BART to perform denoising, which can benefit from both the rich prelearned knowledge of BART and the iterative refining paradigm of DDM.Besides, we also propose the iterative self-prompting strategy to further improve the generation quality.Experimental results on 7 datasets show that our approach can outperform competitive NAR methods, and even surpass autoregressive methods.Our code and data are released at

Large language models (LLMs) are still struggling in aligning with human preference in complex tasks and scenarios.They are prone to overfit into the unexpected patterns or superficial styles in the training data.We conduct an empirical study that only selects the top-10% most updated parameters in LLMs for alignment training, and see improvements in the convergence process and final performance.It indicates the existence of redundant neurons in LLMs for alignment training.To reduce its influence, we propose a low-redundant alignment method named ALLO, focusing on optimizing the most related neurons with the most useful supervised signals.Concretely, we first identify the neurons that are related to the human preference data by a gradient-based strategy, then identify the alignment-related key tokens by reward models for computing loss.Besides, we also decompose the alignment process into the forgetting and learning stages, where we first forget the tokens with unaligned knowledge and then learn aligned knowledge, by updating different ratios of neurons, respectively.Experimental results on 10 datasets have shown the effectiveness of ALLO.Our code and data are available at https://github.com/RUCAIBox/ALLO.

Hallucination detection is a challenging task for large language models (LLMs), and existing studies heavily rely on powerful closed-source LLMs such as GPT-4.In this paper, we propose an autonomous LLM-based agent framework, called HaluAgent, which enables relatively smaller LLMs 1 (e.g., Baichuan2-Chat 7B) to actively select suitable tools for detecting multiple hallucination types such as text, code, and mathematical expression.In HaluAgent, we integrate the LLM, multi-functional toolbox, and design a fine-grained three-stage detection framework along with memory mechanism.To facilitate the effectiveness of HaluAgent, we leverage existing Chinese and English datasets to synthesize detection trajectories for fine-tuning, which endows HaluAgent with the capability for bilingual hallucination detection.Extensive experiments demonstrate that only using 2K samples for tuning LLMs, HaluAgent can perform hallucination detection on various types of tasks and datasets, achieving performance comparable to or even higher than GPT-4 without tool enhancements on both indomain and out-of-domain datasets.We release our dataset and code at https://github.com/ RUCAIBox/HaluAgent.

Considering the limited internal parametric knowledge, retrieval-augmented generation (RAG) has been widely used to extend the knowledge scope of large language models (LLMs).Despite the extensive efforts on RAG research, in existing methods, LLMs cannot precisely assess the relevance of retrieved documents, thus likely leading to misleading or even incorrect utilization of external knowledge (i.e., retrieved documents).To address this issue, in this paper, we propose REAR, a RElevance-Aware Retrievalaugmented approach for open-domain question answering (QA).As the key motivation, we aim to enhance the self-awareness regarding the reliability of external knowledge for LLMs, so as to adaptively utilize external knowledge in RAG systems.Specially, we develop a novel architecture for LLM-based RAG systems, by incorporating a specially designed assessment module that precisely assesses the relevance of retrieved documents.Furthermore, we propose an improved training method based on bigranularity relevance fusion and noise-resistant training.By combining the improvements in both architecture and training, our proposed REAR can better utilize external knowledge by effectively perceiving the relevance of retrieved documents.Experiments on four opendomain QA tasks show that REAR significantly outperforms previous a number of competitive RAG approaches.

Recently, large language models (LLMs) have shown great potential in recommender systems, either improving existing recommendation models or serving as the backbone. However, there exists a large semantic gap between LLMs and recommender systems, since items to be recommended are often indexed by discrete identifiers (item ID) out of the LLM's vocabulary. In essence, LLMs capture language semantics while recommender systems imply collaborative semantics, making it difficult to sufficiently leverage the model capacity of LLMs for recommendation. To address this challenge, in this paper, we propose a new LLM-based recommendation model called LC-Rec, which can better integrate language and collaborative semantics for recommender systems. Our approach can directly generate items from the entire item set for recommendation, without relying on candidate items. Specifically, we make two major contributions in our approach. For item indexing, we design a learning-based vector quantization method with uniform semantic mapping, which can assign meaningful and non-conflicting IDs (called item indices) for items. For alignment tuning, we propose a series of specially designed tuning tasks to enhance the integration of collaborative semantics in LLMs. Our fine-tuning tasks enforce LLMs to deeply integrate language and collaborative semantics (characterized by the learned item indices), so as to achieve an effective adaptation to recommender systems. Extensive experiments demonstrate the effectiveness of our method, showing that our approach can outperform a number of competitive baselines including traditional recommenders and existing LLM-based recommenders. Our code is available at https://github.com/RUCAIBox/LC-Rec/.

Alignment with human preference is a desired property of large language models (LLMs). Currently, the main alignment approach is based on reinforcement learning from human feedback (RLHF). Despite the effectiveness of RLHF, it is intricate to implement and train, thus recent studies explore how to develop alternative alignment approaches based on supervised fine-tuning (SFT). A major limitation of SFT is that it essentially does imitation learning, which can't fully understand what are the expected behaviors. To address this issue, we propose an improved alignment approach named $\textbf{FIGA}$. Different from prior methods, we incorporate fine-grained (i.e., token or phrase level) quality signals that are derived by contrasting good and bad responses. Our approach has made two major contributions. Firstly, we curate a refined alignment dataset that pairs initial responses and the corresponding revised ones. Secondly, we devise a new loss function can leverage fine-grained quailty signals to instruct the learning of LLMs for alignment. Extensive experiments have demonstrated the effectiveness of our approaches by comparing a number of competitive baselines.

Medical vision-language pre-training methods mainly leverage the correspondence between paired medical images and radiological reports. Although multi-view spatial images and temporal sequences of image-report pairs are available in off-the-shelf multi-modal medical datasets, most existing methods have not thoroughly tapped into such extensive supervision signals. In this paper, we introduce the Med-ST framework for fine-grained spatial and temporal modeling to exploit information from multiple spatial views of chest radiographs and temporal historical records. For spatial modeling, Med-ST employs the *Mixture of View Expert (MoVE)* architecture to integrate different visual features from both frontal and lateral views. To achieve a more comprehensive alignment, Med-ST not only establishes the global alignment between whole images and texts but also introduces modality-weighted local alignment between text tokens and spatial regions of images. For temporal modeling, we propose a novel cross-modal bidirectional cycle consistency objective by forward mapping classification (FMC) and reverse mapping regression (RMR). By perceiving temporal information from simple to complex, Med-ST can learn temporal semantics. Experimental results across four distinct tasks demonstrate the effectiveness of Med-ST, especially in temporal classification tasks. Our code and model are available at https://github.com/SVT-Yang/MedST.

Feature interaction learning (FIL) focuses on capturing the complex relationships among multiple features for building predictive models, which is widely used in real-world tasks. Despite the research progress, existing FIL methods suffer from two major limitations. Firstly, they mainly model the feature interactions within a bounded order (e.g., small integer order) due to the exponential growth of the interaction terms. Secondly, the interaction order of each feature is often independently learned, which lacks the flexibility to capture the feature dependencies in varying contexts.

Reciprocal recommender systems~(RRS), conducting bilateral recommendations between two involved parties, have gained increasing attention for enhancing matching efficiency. However, the majority of existing methods in the literature still reuse conventional ranking metrics to separately assess the performance on each side of the recommendation process. These methods overlook the fact that the ranking outcomes of both sides collectively influence the effectiveness of the RRS, neglecting the necessity of a more holistic evaluation and a capable systemic solution.

Tianyi Tang, Hongyuan Lu, Yuchen Jiang, Haoyang Huang, Dongdong Zhang, Xin Zhao, Tom Kocmi, Furu Wei. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Mathematical reasoning is an important capability of large language models~(LLMs) for real-world applications. To enhance this capability, existing work either collects large-scale math-related texts for pre-training, or relies on stronger LLMs (\eg GPT-4) to synthesize massive math problems. Both types of work generally lead to large costs in training or synthesis. To reduce the cost, based on open-source available texts, we propose an efficient way that trains a small LLM for math problem synthesis, to efficiently generate sufficient high-quality pre-training data. To achieve it, we create a dataset using GPT-4 to distill its data synthesis capability into the small LLM. Concretely, we craft a set of prompts based on human education stages to guide GPT-4, to synthesize problems covering diverse math knowledge and difficulty levels. Besides, we adopt the gradient-based influence estimation method to select the most valuable math-related texts. The both are fed into GPT-4 for creating the knowledge distillation dataset to train the small LLM. We leverage it to synthesize 6 million math problems for pre-training our JiuZhang3.0 model. The whole process only needs to invoke GPT-4 API 9.3k times and use 4.6B data for training. Experimental results have shown that JiuZhang3.0 achieves state-of-the-art performance on several mathematical reasoning datasets, under both natural language reasoning and tool manipulation settings. Our code and data will be publicly released in \url{https://github.com/RUCAIBox/JiuZhang3.0}.

Transformer-based large language models (LLMs) typically have a limited context window, resulting in significant performance degradation when processing text beyond the length of the context window. Extensive studies have been proposed to extend the context window and achieve length extrapolation of LLMs, but there is still a lack of in-depth interpretation of these approaches. In this study, we explore the positional information within and beyond the context window for deciphering the underlying mechanism of LLMs. By using a mean-based decomposition method, we disentangle positional vectors from hidden states of LLMs and analyze their formation and effect on attention. Furthermore, when texts exceed the context window, we analyze the change of positional vectors in two settings, i.e., direct extrapolation and context window extension. Based on our findings, we design two training-free context window extension methods, positional vector replacement and attention window extension. Experimental results show that our methods can effectively extend the context window length.

At present, most recommender systems involve two stakeholders, providers and customers. Apart from maximizing the recommendation accuracy, the fairness issue for both sides should also be considered. Most of previous studies try to improve two-sided fairness with post-processing algorithms or fairness-aware loss constraints, which are highly dependent on the heuristic adjustments without respect to the optimization goal of accuracy. In contrast, we propose a novel training framework, adaptive weighting towards two-sided fairness-aware recommendation (named Ada2Fair), which lies in the extension of the accuracy-focused objective to a controllable preference learning loss over the interaction data. Specifically, we adjust the optimization scale of an interaction sample with an adaptive weight generator, and estimate the two-sided fairness-aware weights within model training. During the training process, the recommender is trained with two-sided fairness-aware weights to boost the utility of niche providers and inactive customers in a unified way. Extensive experiments on three public datasets verify the effectiveness of Ada2Fair, which can achieve Pareto efficiency in two-sided fairness-aware recommendation.

Scaling of neural networks has recently shown great potential to improve the model capacity in various fields. Specifically, model performance has a power-law relationship with model size or data size, which provides important guidance for the development of large-scale models. However, there is still limited understanding on the scaling effect of user behavior models in recommender systems, where the unique data characteristics (e.g., data scarcity and sparsity) pose new challenges in recommendation tasks.

The ability of the foundation models heavily relies on large-scale, diverse, and high-quality pretraining data. In order to improve data quality, researchers and practitioners often have to manually curate datasets from difference sources and develop dedicated data cleansing pipeline for each data repository. Lacking a unified data processing framework, this process is repetitive and cumbersome. To mitigate this issue, we propose a data processing framework that integrates a Processing Module which consists of a series of operators at different granularity levels, and an Analyzing Module which supports probing and evaluation of the refined data. The proposed framework is easy to use and highly flexible. In this demo paper, we first introduce how to use this framework with some example use cases and then demonstrate its effectiveness in improving the data quality with an automated evaluation with ChatGPT and an end-to-end evaluation in pretraining the GPT-2 model. The code and demonstration video are accessible on GitHub.

To capture user preference, transformer models have been widely applied to model sequential user behavior data. The core of transformer architecture lies in the self-attention mechanism, which computes the pairwise attention scores in a sequence. Due to the permutation-equivariant nature, positional encoding is used to enhance the attention between token representations. In this setting, the pairwise attention scores can be derived by both semantic difference and positional difference. However, prior studies often model the two kinds of difference measurements in different ways, which potentially limits the expressive capacity of sequence modeling.

Abstract Autonomous agents have long been a research focus in academic and industry communities. Previous research often focuses on training agents with limited knowledge within isolated environments, which diverges significantly from human learning processes, and makes the agents hard to achieve human-like decisions. Recently, through the acquisition of vast amounts of Web knowledge, large language models (LLMs) have shown potential in human-level intelligence, leading to a surge in research on LLM-based autonomous agents. In this paper, we present a comprehensive survey of these studies, delivering a systematic review of LLM-based autonomous agents from a holistic perspective. We first discuss the construction of LLM-based autonomous agents, proposing a unified framework that encompasses much of previous work. Then, we present a overview of the diverse applications of LLM-based autonomous agents in social science, natural science, and engineering. Finally, we delve into the evaluation strategies commonly used for LLM-based autonomous agents. Based on the previous studies, we also present several challenges and future directions in this field.

As people inevitably interact with items across multiple domains or various platforms, cross-domain recommendation (CDR) has gained increasing attention. However, the rising privacy concerns limit the practical applications of existing CDR models, since they assume that full or partial data are accessible among different domains. Recent studies on privacy-aware CDR models neglect the heterogeneity from multiple-domain data and fail to achieve consistent improvements in cross-domain recommendation; thus, it remains a challenging task to conduct effective CDR in a privacy-preserving way. In this article, we propose a novel, as far as we know, federated graph learning approach for Privacy-Preserving Cross-Domain Recommendation (PPCDR) to capture users’ preferences based on distributed multi-domain data and improve recommendation performance for all domains without privacy leakage. The main idea of PPCDR is to model both global preference among multiple domains and local preference at a specific domain for a given user, which characterizes the user’s shared and domain-specific tastes toward the items for interaction. Specifically, in the private update process of PPCDR, we design a graph transfer module for each domain to fuse global and local user preferences and update them based on local domain data. In the federated update process, through applying the local differential privacy technique for privacy-preserving, we collaboratively learn global user preferences based on multi-domain data and adapt these global preferences to heterogeneous domain data via personalized aggregation. In this way, PPCDR can effectively approximate the multi-domain training process that directly shares local interaction data in a privacy-preserving way. Extensive experiments on three CDR datasets demonstrate that PPCDR consistently outperforms competitive single- and cross-domain baselines and effectively protects domain privacy.

Text retrieval is a long-standing research topic on information seeking, where a system is required to return relevant information resources to user’s queries in natural language. From heuristic-based retrieval methods to learning-based ranking functions, the underlying retrieval models have been continually evolved with the ever-lasting technical innovation. To design effective retrieval models, a key point lies in how to learn text representations and model the relevance matching. The recent success of pretrained language models (PLM) sheds light on developing more capable text-retrieval approaches by leveraging the excellent modeling capacity of PLMs. With powerful PLMs, we can effectively learn the semantic representations of queries and texts in the latent representation space, and further construct the semantic matching function between the dense vectors for relevance modeling. Such a retrieval approach is called dense retrieval , since it employs dense vectors to represent the texts. Considering the rapid progress on dense retrieval, this survey systematically reviews the recent progress on PLM-based dense retrieval. Different from previous surveys on dense retrieval, we take a new perspective to organize the related studies by four major aspects, including architecture, training, indexing and integration, and thoroughly summarize the mainstream techniques for each aspect. We extensively collect the recent advances on this topic, and include 300+ reference papers. To support our survey, we create a website for providing useful resources, and release a code repository for dense retrieval. This survey aims to provide a comprehensive, practical reference focused on the major progress for dense text retrieval.

As a core technology of Intelligent Transportation System, traffic flow prediction has a wide range of applications. The fundamental challenge in traffic flow prediction is to effectively model the complex spatial-temporal dependencies in traffic data. Spatial-temporal Graph Neural Network (GNN) models have emerged as one of the most promising methods to solve this problem. However, GNN-based models have three major limitations for traffic prediction: i) Most methods model spatial dependencies in a static manner, which limits the ability to learn dynamic urban traffic patterns; ii) Most methods only consider short-range spatial information and are unable to capture long-range spatial dependencies; iii) These methods ignore the fact that the propagation of traffic conditions between locations has a time delay in traffic systems. To this end, we propose a novel Propagation Delay-aware dynamic long-range transFormer, namely PDFormer, for accurate traffic flow prediction. Specifically, we design a spatial self-attention module to capture the dynamic spatial dependencies. Then, two graph masking matrices are introduced to highlight spatial dependencies from short- and long-range views. Moreover, a traffic delay-aware feature transformation module is proposed to empower PDFormer with the capability of explicitly modeling the time delay of spatial information propagation. Extensive experimental results on six real-world public traffic datasets show that our method can not only achieve state-of-the-art performance but also exhibit competitive computational efficiency. Moreover, we visualize the learned spatial-temporal attention map to make our model highly interpretable.

Simulating the human mobility and generating large-scale trajectories are of great use in many real-world applications, such as urban planning, epidemic spreading analysis, and geographic privacy protect. Although many previous works have studied the problem of trajectory generation, the continuity of the generated trajectories has been neglected, which makes these methods useless for practical urban simulation scenarios. To solve this problem, we propose a novel two-stage generative adversarial framework to generate the continuous trajectory on the road network, namely TS-TrajGen, which efficiently integrates prior domain knowledge of human mobility with model-free learning paradigm. Specifically, we build the generator under the human mobility hypothesis of the A* algorithm to learn the human mobility behavior. For the discriminator, we combine the sequential reward with the mobility yaw reward to enhance the effectiveness of the generator. Finally, we propose a novel two-stage generation process to overcome the weak point of the existing stochastic generation process. Extensive experiments on two real-world datasets and two case studies demonstrate that our framework yields significant improvements over the state-of-the-art methods.

By scaling the model size, large pre-trained language models (PLMs) have shown remarkable performance in various natural language processing tasks, mostly outperforming small PLMs by a large margin.However, due to the high computational cost, the huge number of parameters also restricts the applicability of large PLMs in real-world systems.In this paper, we focus on scaling up the parameters of PLMs only during fine-tuning, to benefit from the over-parameterization, while without increasing the inference latency.Given a relatively small PLM, we over-parameterize it by employing a matrix product operator, an efficient and almost lossless decomposition method to factorize its contained parameter matrices into a set of higher-dimensional tensors.Considering the efficiency, we further propose both static and dynamic strategies to select the most important parameter matrices for over-parameterization.Extensive experiments have demonstrated that our approach can significantly boost the finetuning performance of small PLMs and even help small PLMs outperform 3× parameterized larger ones.

Although pre-trained language models (PLMs) have shown impressive performance by textonly self-supervised training, they are found lack of visual semantics or commonsense.Existing solutions often rely on explicit images for visual knowledge augmentation (requiring time-consuming retrieval or generation), and they also conduct the augmentation for the whole input text, without considering whether it is actually needed in specific inputs or tasks.To address these issues, we propose a novel Visually-Augmented fine-tuning approach that can be generally applied to various PLMs or NLP tasks, Without using any retrieved or generated Images, namely VAWI.Experimental results show that our approach can consistently improve the performance of BERT, RoBERTa, BART, and T5 at different scales, and outperform several competitive baselines on ten tasks.Our codes and data are publicly available at https://github.com/RUCAIBox/VAWI.

Recently, model-based retrieval has emerged as a new paradigm in text retrieval that discards the index in the traditional retrieval model and instead memorizes the candidate corpora using model parameters.This design employs a sequence-to-sequence paradigm to generate document identifiers, which enables the complete capture of the relevance between queries and documents and simplifies the classic indexretrieval-rerank pipeline.Despite its attractive qualities, there remain several major challenges in model-based retrieval, including the discrepancy between pre-training and fine-tuning, and the discrepancy between training and inference.To deal with the above challenges, we propose a novel two-stage model-based retrieval approach called TOME, which makes two major technical contributions, including the utilization of tokenized URLs as identifiers and the design of a two-stage generation architecture.We also propose a number of training strategies to deal with the training difficulty as the corpus size increases.Extensive experiments and analysis on MS MARCO and Natural Questions demonstrate the effectiveness of our proposed approach, and we investigate the scaling laws of TOME by examining various influencing factors.

People often imagine relevant scenes to aid in the writing process.In this work, we aim to utilize visual information for composition in the same manner as humans.We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration.First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts.Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way.Finally, our imagination is dynamic, and we conduct synthesis for each sentence rather than generate only one image for an entire paragraph.Technically, we propose a novel plug-and-play fusion layer to obtain visually-augmented representations for each text.Our vision-text fusion layer is compatible with Transformerbased architecture.We have conducted extensive experiments on four generation tasks using BART and T5, and the automatic results and human evaluation demonstrate the effectiveness of our proposed method.We will release the code, model, and data at the link: https://github.com/RUCAIBox/LIVE.

Within online platforms, it is critical to capture the dynamic user preference from the sequential interaction behaviors for making accurate recommendation over time. Recently, significant progress has been made in sequential recommendation with deep learning. However, existing neural sequential recommender often suffer from the data sparsity issue in real-world applications.

Multi-scenario ad ranking aims at leveraging the data from multiple domains or channels for training a unified ranking model for improving the performance at each individual scenario. Although the research on this task has made important progress, it still lacks the consideration of cross-scenario relations, thus leading to limitation in learning capability and difficulty in interrelation modeling.

Nowadays, contrastive self-supervised learning has been widely incorporated into sequential recommender systems. However, most existing contrastive sequential recommender systems simply emphasize the overall information of interaction sequences, thereby neglecting the special periodic patterns of user behavior. In this study, we propose that users exhibit emanative periodicity towards a group of correlated items, i.e., user behavior follow a certain periodic pattern while their interests may shift from one item to other related items over time. In light of this observation, we present a hierarchical contrastive learning framework to model EmAnative periodicity for SEquential Recommendation (referred to as EASE). Specifically, we design dual-channel contrastive strategy from the perspective of correlation and periodicity to capture emanative periodic patterns. Furthermore, we extend the traditional binary contrastive loss with hierarchical constraint to handle hierarchical contrastive samples, thus preserving the inherent hierarchical information of correlation and periodicity. Comprehensive experiments conducted on five datasets substantiate the effectiveness of our proposed EASE in improving sequential recommendation.

In this paper, we aim to improve the reasoning ability of large language models (LLMs) over structured data in a unified way.Inspired by the studies on tool augmentation for LLMs, we develop an Iterative Reading-then-Reasoning (IRR) framework to solve question answering tasks based on structured data, called StructGPT.In this framework, we construct the specialized interfaces to collect relevant evidence from structured data (i.e., reading), and let LLMs concentrate on the reasoning task based on the collected information (i.e., reasoning).Specially, we propose an invokinglinearization-generation procedure to support LLMs in reasoning on the structured data with the help of the interfaces.By iterating this procedure with provided interfaces, our approach can gradually approach the target answers to a given query.Experiments conducted on three types of structured data show that StructGPT greatly improves the performance of LLMs, under the few-shot and zero-shot settings.Our codes and data are publicly available at https://github.com/RUCAIBox/StructGPT.

Question Answering over Knowledge Graph (KGQA) aims to seek answer entities for the natural language question from a largescale Knowledge Graph (KG).To better perform reasoning on KG, recent work typically adopts a pre-trained language model (PLM) to model the question, and a graph neural network (GNN) based module to perform multihop reasoning on the KG.Despite the effectiveness, due to the divergence in model architecture, the PLM and GNN are not closely integrated, limiting the knowledge sharing and finegrained feature interactions.To solve it, we aim to simplify the above two-module approach, and develop a more capable PLM that can directly support subgraph reasoning for KGQA, namely ReasoningLM.In our approach, we propose a subgraph-aware self-attention mechanism to imitate the GNN for performing structured reasoning, and also adopt an adaptation tuning strategy to adapt the model parameters with 20,000 subgraphs with synthesized questions.After adaptation, the PLM can be parameter-efficient fine-tuned on downstream tasks.Experiments show that ReasoningLM surpasses state-of-the-art models by a large margin, even with fewer updated parameters and less training data.Our codes and data are publicly available at https://github.com/ RUCAIBox/ReasoningLM.

Large language models (LLMs), such as Chat-GPT, are prone to generate hallucinations, i.e., content that conflicts with the source or cannot be verified by the factual knowledge.To understand what types of content and to which extent LLMs are apt to hallucinate, we introduce the Hallucination Evaluation benchmark for Large Language Models (HaluEval), a large collection of generated and human-annotated hallucinated samples for evaluating the performance of LLMs in recognizing hallucination.To generate these samples automatically, we propose a two-stage framework, i.e., samplingthen-filtering.Besides, we hire some human labelers to annotate the hallucinations in Chat-GPT responses.The empirical results suggest that ChatGPT is likely to generate hallucinated content related to specific topics by fabricating unverifiable information (i.e., about 19.5% responses).Moreover, existing LLMs face great challenges in recognizing the hallucinations in texts.However, our experiments also prove that providing external knowledge or adding reasoning steps can help LLMs recognize hallucinations.Our benchmark can be accessed at https://github.com/RUCAIBox/HaluEval.

Inspired by the superior language abilities of large language models (LLM), large visionlanguage models (LVLM) have been recently proposed by integrating powerful LLMs for improving the performance on complex multimodal tasks.Despite the promising progress on LVLMs, we find that they suffer from object hallucinations, i.e., they tend to generate objects inconsistent with the target images in the descriptions.To investigate it, this work presents the first systematic study on object hallucination of LVLMs.We conduct the evaluation experiments on several representative LVLMs, and show that they mostly suffer from severe object hallucination issues.We further discuss that the visual instructions may influence the hallucination, and find that: objects that frequently appear in the visual instructions or co-occur with the image objects are obviously prone to be hallucinated by LVLMs.Besides, we further design a polling-based query method called POPE for better evaluation of object hallucination.Experiment results show that our POPE can evaluate object hallucination in a more stable and flexible way.

The recent success of large language models (LLMs) has shown great potential to develop more powerful conversational recommender systems (CRSs), which rely on natural language conversations to satisfy user needs.In this paper, we embark on an investigation into the utilization of ChatGPT for CRSs, revealing the inadequacy of the existing evaluation protocol.It might overemphasize the matching with ground-truth items annotated by humans while neglecting the interactive nature of CRSs.To overcome the limitation, we further propose an interactive Evaluation approach based on LLMs, named iEvaLM, which harnesses LLM-based user simulators.Our evaluation approach can simulate various system-user interaction scenarios.Through the experiments on two public CRS datasets, we demonstrate notable improvements compared to the prevailing evaluation protocol.Furthermore, we emphasize the evaluation of explainability, and Chat-GPT showcases persuasive explanation generation for its recommendations.Our study contributes to a deeper comprehension of the untapped potential of LLMs for CRSs and provides a more flexible and realistic evaluation approach for future research about LLMbased CRSs.The code is available at https: //github.com/RUCAIBox/iEvaLM-CRS.

Non-autoregressive (NAR) text generation has attracted much attention in the field of natural language processing, which greatly reduces the inference latency but has to sacrifice the generation accuracy. Recently, diffusion models, a class of latent variable generative models, have been introduced into NAR text generation, showing an improved text generation quality. In this survey, we review the recent progress in diffusion models for NAR text generation. As the background, we first present the general definition of diffusion models and the text diffusion models, and then discuss their merits for NAR generation. As the core content, we further introduce two mainstream diffusion models in existing work of text diffusion, and review the key designs of the diffusion process. Moreover, we discuss the utilization of pre-trained language models (PLMs) for text diffusion models and introduce optimization techniques for text data. Finally, we discuss several promising directions and conclude this paper. Our survey aims to provide researchers with a systematic reference of related research on text diffusion models for NAR generation. We also demonstrate our collection of text diffusion models at https://github.com/RUCAIBox/Awesome-Text-Diffusion-Models.

Conversational recommender systems~(CRSs) aim to provide recommendation services via natural language conversations. Although a number of approaches have been proposed for developing capable CRSs, they typically rely on sufficient training data for training. Since it is difficult to annotate recommendation-oriented dialogue datasets, existing CRS approaches often suffer from the issue of insufficient training due to the scarcity of training data.

Although pre-trained language models~(PLMs) have recently advanced the research progress in mathematical reasoning, they are not specially designed as a capable multi-task solver, suffering from high cost for multi-task deployment (e.g. a model copy for a task) and inferior performance on complex mathematical problems in practical applications. To address these issues, we propose JiuZhang 2.0, a unified Chinese PLM specially for multi-task mathematical problem solving. Our idea is to maintain a moderate-sized model and employ the cross-task knowledge sharing to improve the model capacity in a multi-task setting. Specially, we construct a Mixture-of-Experts (MoE) architecture for modeling mathematical text, to capture the common mathematical knowledge across tasks. For optimizing the MoE architecture, we design multi-task continual pre-training and multi-task fine-tuning strategies for multi-task adaptation. These training strategies can effectively decompose the knowledge from the task data and establish the cross-task sharing via expert networks. To further improve the general capacity of solving different complex tasks, we leverage large language models (LLMs) as complementary models to iteratively refine the generated solution by our PLM, via in-context learning. Extensive experiments have demonstrated the effectiveness of our model.

To facilitate the research on intelligent and human-like chatbots with multi-modal context, we introduce a new video-based multi-modal dialogue dataset, called TikTalk. We collect 38K videos from a popular video-sharing platform, along with 367K conversations posted by users beneath them. Users engage in spontaneous conversations based on their multi-modal experiences from watching videos, which helps recreate real-world chitchat context. Compared to previous multi-modal dialogue datasets, the richer context types in TikTalk lead to more diverse conversations, but also increase the difficulty in capturing human interests from intricate multi-modal information to generate personalized responses. Moreover, external knowledge is more frequently evoked in our dataset. These facts reveal new challenges for multi-modal dialogue models. We quantitatively demonstrate the characteristics of TikTalk, propose a video-based multi-modal chitchat task, and evaluate several dialogue baselines. Experimental results indicate that the models incorporating large language models (LLM) can generate more diverse responses, while the model utilizing knowledge graphs to introduce external knowledge performs the best overall. Furthermore, no existing model can solve all the above challenges well. There is still a large room for future improvements, even for LLM with visual extensions. Our dataset is available at https://ruc-aimind.github.io/projects/TikTalk/.

Explainable recommendation has attracted much attention from the industry and academic communities. It has shown great potential to improve the recommendation persuasiveness, informativeness and user satisfaction. In the past few years, while a lot of promising explainable recommender models have been proposed, the datasets used to evaluate them still suffer from several limitations, for example, the explanation ground truths are not labeled by the real users, the explanations are mostly single-modal and around only one aspect. To bridge these gaps, in this paper, we build a new explainable recommendation dataset, which, to our knowledge, is the first contribution that provides a large amount of real user labeled multi-modal and multi-aspect explaination ground truths. In specific, we firstly develop a video recommendation platform, where a series of questions around the recommendation explainability are carefully designed. Then, we recruit about 3000 high-quality labelers with different backgrounds to use the system, and collect their behaviors and feedback to our questions. In this paper, we detail the construction process of our dataset and also provide extensive analysis on its characteristics. In addition, we develop a library, where ten well-known explainable recommender models are implemented in a unified framework. Based on this library, we build several benchmarks for different explainable recommendation tasks. At last, we present many new opportunities brought by our dataset, which are expected to promote the field of explainable recommendation. Our dataset, library and the related documents have been released at https://reasoner2023.github.io/.

Chain-of-thought prompting (CoT) and tool augmentation have been validated in recent work as effective practices for improving large language models (LLMs) to perform step-by-step reasoning on complex math-related tasks.However, most existing math reasoning datasets may not be able to fully evaluate and analyze the ability of LLMs in manipulating tools and performing reasoning, as they often only require very few invocations of tools or miss annotations for evaluating intermediate reasoning steps, thus supporting only outcome evaluation.To address the issue, we construct CARP, a new Chinese dataset consisting of 4,886 computation-intensive algebra problems with formulated annotations on intermediate steps, facilitating the evaluation of the intermediate reasoning process.In CARP, we test four LLMs with CoT prompting, and find that they are all prone to make mistakes at the early steps of the solution, leading to incorrect answers.Based on this finding, we propose a new approach that can facilitate the deliberation on reasoning steps with tool interfaces, namely DELI.In DELI, we first initialize a step-by-step solution based on retrieved exemplars, then iterate two deliberation procedures that check and refine the intermediate steps of the generated solution, from both tool manipulation and natural language reasoning perspectives, until solutions converge or the maximum iteration is achieved.Experimental results on CARP and six other datasets show that the proposed DELI mostly outperforms competitive baselines, and can further boost the performance of existing CoT methods.Our data and code are available at https://github.com/RUCAIBox/CARP.

Next-basket Recommendation (NBR) refers to the task of predicting a set of items that a user will purchase in the next basket. However, most of existing works merely focus on the correlations between user preferences and predicted items, ignoring the essential correlations among items in the next basket, which often results in over-homogenization of predicted items. In this work, we presents a Generative next-basket Recommendation model (GenRec), a novel NBR paradigm that generates the recommended items one by one to form the next basket via an autoregressive decoder. This generative NBR paradigm contributes to capturing and considering item correlations inside each baskets in both training and serving. Moreover, we jointly consider user’s both item- and basket-level contextual information to better capture user’s multi-granularity preferences. Extensive experiments on three real-world datasets demonstrate the effectiveness of our model.

Conversational recommender systems (CRS) aim to provide the recommendation service via natural language conversations. To develop an effective CRS, high-quality CRS datasets are very crucial. However, existing CRS datasets suffer from the long-tail issue, i.e., a large proportion of items are rarely (or even never) mentioned in the conversations, which are called long-tail items. As a result, the CRSs trained on these datasets tend to recommend frequent items, and the diversity of the recommended items would be largely reduced, making users easier to get bored.

Reciprocal recommender system (RRS), considering a two-way matching between two parties, has been widely applied in online platforms like online dating and recruitment. Existing RRS models mainly capture static user preferences, which have neglected the evolving user tastes and the dynamic matching relation between the two parties. Although dynamic user modeling has been well-studied in sequential recommender systems, existing solutions are developed in a user-oriented manner. Therefore, it is non-trivial to adapt sequential recommendation algorithms to reciprocal recommendation. In this paper, we formulate RRS as a distinctive sequence matching task, and further propose a new approach ReSeq for RRS, which is short for Reciprocal Sequential recommendation. To capture dual-perspective matching, we propose to learn fine-grained sequence similarities by co-attention mechanism across different time steps. Further, to improve the inference efficiency, we introduce the self-distillation technique to distill knowledge from the fine-grained matching module into the more efficient student module. In the deployment stage, only the efficient student module is used, greatly speeding up the similarity computation. Extensive experiments on five real-world datasets from two scenarios demonstrate the effectiveness and efficiency of the proposed method. Our code is available at https://github.com/RUCAIBox/ReSeq/.

Learning effective high-order feature interactions is very crucial in the CTR prediction task. However, it is very time-consuming to calculate high-order feature interactions with massive features in online e-commerce platforms. Most existing methods manually design a maximal order and further filter out the useless interactions from them. Although they reduce the high computational costs caused by the exponential growth of high-order feature combinations, they still suffer from the degradation of model capability due to the suboptimal learning of the restricted feature orders. The solution to maintain the model capability and meanwhile keep it efficient is a technical challenge, which has not been adequately addressed. To address this issue, we propose an adaptive feature interaction learning model, named as EulerNet, in which the feature interactions are learned in a complex vector space by conducting space mapping according to Euler's formula. EulerNet converts the exponential powers of feature interactions into simple linear combinations of the modulus and phase of the complex features, making it possible to adaptively learn the high-order feature interactions in an efficient way. Furthermore, EulerNet incorporates the implicit and explicit feature interactions into a unified architecture, which achieves the mutual enhancement and largely boosts the model capabilities. Such a network can be fully learned from data, with no need of pre-designed form or order for feature interactions. Extensive experiments conducted on three public datasets have demonstrated the effectiveness and efficiency of our approach. Our code is available at: https://github.com/RUCAIBox/EulerNet.

In recent years, the reproducibility of recommendation models has become a severe concern in recommender systems. In light of this challenge, we have previously released a unified, comprehensive and efficient recommendation library called RecBole, attracting much attention from the research community. With the increasing number of users, we have received a number of suggestions and update requests. This motivates us to make further improvements on our library, so as to meet the user requirements and contribute to the research community. In this paper, we present a significant update of RecBole, making it more user-friendly and easy-to-use as a comprehensive benchmark library for recommendation. More specifically, the highlights of this update are summarized as: (1) we include more benchmark models and datasets, improve the benchmark framework in terms of data processing, training and evaluation, and release reproducible configurations to benchmark the recommendation models; (2) we upgrade the user friendliness of our library by providing more detailed documentation and well-organized frequently asked questions, and (3) we propose several development guidelines for the open-source library developers. These extensions make it much easier to reproduce the benchmark results and stay up-to-date with the recent advances on recommender systems. Our update is released at the link: https://github.com/RUCAIBox/RecBole.

With the growth of high-dimensional sparse data in web-scale recommender systems, the computational cost to learn high-order feature interaction in CTR prediction task largely increases, which limits the use of high-order interaction models in real industrial applications. Some recent knowledge distillation based methods transfer knowledge from complex teacher models to shallow student models for accelerating the online model inference. However, they suffer from the degradation of model accuracy in knowledge distillation process. It is challenging to balance the efficiency and effectiveness of the shallow student models. To address this problem, we propose a Directed Acyclic Graph Factorization Machine (KD-DAGFM) to learn the high-order feature interactions from existing complex interaction models for CTR prediction via Knowledge Distillation. The proposed lightweight student model DAGFM can learn arbitrary explicit feature interactions from teacher networks, which achieves approximately lossless performance and is proved by a dynamic programming algorithm. Besides, an improved general model KD-DAGFM+ is shown to be effective in distilling both explicit and implicit feature interactions from any complex teacher model. Extensive experiments are conducted on four real-world datasets, including a large-scale industrial dataset from WeChat platform with billions of feature dimensions. KD-DAGFM achieves the best performance with less than 21.5% FLOPs of the state-of-the-art method on both online and offline experiments, showing the superiority of DAGFM to deal with the industrial scale data in CTR prediction task.

Recently, the generality of natural language text has been leveraged to develop transferable recommender systems. The basic idea is to employ pre-trained language models (PLM) to encode item text into item representations. Despite the promising transferability, the binding between item text and item representations might be too tight, leading to potential problems such as over-emphasizing the effect of text features and exaggerating the negative impact of domain gap. To address this issue, this paper proposes VQ-Rec, a novel approach to learning Vector-Quantized item representations for transferable sequential Recommenders. The main novelty of our approach lies in the new item representation scheme: it first maps item text into a vector of discrete indices (called item code), and then employs these indices to lookup the code embedding table for deriving item representations. Such a scheme can be denoted as “text ⟹ code ⟹ representation”. Based on this representation scheme, we further propose an enhanced contrastive pre-training approach, using semi-synthetic and mixed-domain code representations as hard negatives. Furthermore, we design a new cross-domain fine-tuning method based on a differentiable permutation-based network. Extensive experiments conducted on six public benchmarks demonstrate the effectiveness of the proposed approach, in both cross-domain and cross-platform settings. Code and pre-trained model are available at: https://github.com/RUCAIBox/VQ-Rec.

Recently, contrastive learning has been shown effective in fine-tuning pre-trained language models (PLM) to learn sentence representations, which incorporates perturbations into unlabeled sentences to augment semantically related positive examples for training. However, previous works mostly adopt heuristic perturbation methods that are independent of the sentence representations. Since the perturbations are unaware of the goal or process of sentence representation learning during training, it is likely to lead to sub-optimal augmentations for conducting constrative learning. To address this issue, we propose a new frameworkL2P-CSRthat adopts a learnable perturbation strategy for improving contrastive learning of sentence representations. In our L2P-CSR, we design a safer perturbation mechanism that only weakens the influence of tokens and features on the sentence representation, which avoids dramatically changing the semantics of the sentence representations. Besides, we devise a gradient-based algorithm to generate adaptive perturbations specially for the dynamically updated sentence representation during training. Such a way is more capable of augmenting high-quality examples that guide the sentence representation learning. Extensive experiments on diverse sentence-related tasks show that our approach outperforms competitive baselines.

Sequential recommendation has recently attracted increasing attention from the industry and academic communities. While previous models have achieved remarkable successes, an important problem may still hinder their performances, that is, the sparsity of the real-world data. In this paper, we propose a novel counterfactual data augmentation framework to alleviate the problem of data sparsity. In specific, our framework contains a sampler model and an anchor model. The sampler model aims to generate high-quality user behavior sequences, while the anchor model is trained based on the original and new generated samples, and leveraged to provide the final recommendation list. To implement the sampler model, we first design four types of heuristic methods based on either random or frequency-based strategies. And then, to improve the quality of the generated sequences, we propose two learning-based samplers by discovering the decision boundaries or increasing the sample informativeness. At last, we build an RL based model to automatically determine where to edit the history behaviors and how many items should be replaced. Considering that the sampler model can be imperfect, we, at last, analyze the influence of the noisy information contained in the generated sequences on the anchor model in theory, and design a simple but effective method to better serve the anchor model. We conduct extensive experiments to demonstrate the effectiveness of our model.

Knowledge base question answering (KBQA) aims to answer a question over a knowledge base (KB). Early studies mainly focused on answering simple questions over KBs and achieved great success. However, their performances on complex questions are still far from satisfactory. Therefore, in recent years, researchers propose a large number of novel methods, which looked into the challenges of answering complex questions. In this survey, we review recent advances in KBQA with the focus on solving complex questions, which usually contain multiple subjects, express compound relations, or involve numerical operations. In detail, we begin with introducing the complex KBQA task and relevant background. Then, we present two mainstream categories of methods for complex KBQA, namely semantic parsing-based (SP-based) methods and information retrieval-based (IR-based) methods. Specifically, we illustrate their procedures with flow designs and discuss their difference and similarity. Next, we summarize the challenges that these two categories of methods encounter when answering complex questions, and explicate advanced solutions as well as techniques used in existing work. After that, we discuss the potential impact of pre-trained language models (PLMs) on complex KBQA. To help readers catch up with SOTA methods, we also provide a comprehensive evaluation and resource about complex KBQA task. Finally, we conclude and discuss several promising directions related to complex KBQA for future research.

To characterize complex and heterogeneous side information in recommender systems, the heterogeneous information network (HIN) has shown superior performance and attracted much research attention. In HIN, the rich entities, relations, and paths can be utilized to model the correlations of users and items; such a task setting is often calledHIN-based recommendation. Although HIN provides a general approach to modeling rich side information, it lacks special consideration on the goal of the recommendation task. The aggregated context from the heterogeneous graph is likely to incorporate irrelevant information, and the learned representations are not specifically optimized according to the recommendation task. Therefore, there is a need to rethink how to leverage the useful information from HIN to accomplish the recommendation task. To address the above issues, we propose a Curriculum pre-training based HEterogeneous Subgraph Transformer (calledCHEST) with newdata characterization,representation model,andlearning algorithm. Specifically, we consider extracting useful information from HIN to compose the interaction-specific heterogeneous subgraph, containing highly relevant context information for recommendation. Then, we capture the rich semantics (e.g., graph structure and path semantics) within the subgraph via a heterogeneous subgraph Transformer, where we encode the subgraph into multi-slot sequence representations. Besides, we design a curriculum pre-training strategy to provide an elementary-to-advanced learning process. The elementary course focuses on capturing local context information within the subgraph, and the advanced course aims to learn global context information. In this way, we gradually capture useful semantic information from HIN for modeling user-item interactions. Extensive experiments conducted on four real-world datasets demonstrate the superiority of our proposed method over a number of competitive baselines, especially when only limited training data is available.

In recommender systems, top- N recommendation is an important task with implicit feedback data. Although the recent success of deep learning largely pushes forward the research on top- N recommendation, there are increasing concerns on appropriate evaluation of recommendation algorithms. It therefore is important to study how recommendation algorithms can be reliably evaluated and thoroughly verified. This work presents a large-scale, systematic study on six important factors from three aspects for evaluating recommender systems. We carefully select 12 top- N recommendation algorithms and eight recommendation datasets. Our experiments are carefully designed and extensively conducted with these algorithms and datasets. In particular, all the experiments in our work are implemented based on an open sourced recommendation library, Recbole [ 139 ], which ensures the reproducibility and reliability of our results. Based on the large-scale experiments and detailed analysis, we derive several key findings on the experimental settings for evaluating recommender systems. Our findings show that some settings can lead to substantial or significant differences in performance ranking of the compared algorithms. In response to recent evaluation concerns, we also provide several suggested settings that are specially important for performance comparison.

Sequential recommendation models aim to predict the interested items to a user based on his historical behaviors. To train sequential recommenders, implicit feedback data is widely adopted since it is easier to obtain than explicit feedback data. In the setting of implicit feedback, a user’s historical behaviors can be characterized as a chronologically ordered sequence of interacted items. From a perspective of machine learning, the historical interaction sequence and the recommended items can be considered as context and label , respectively, which are usually in one-hot representations in the recommendation models. However, due to the discrete nature, one-hot representations are hard to sufficiently reflect the underlying user preference, and might also contain noise from implicit feedback that will mislead the model training. To solve these issues, we propose a general optimization framework, Multi-View Smoothness (MVS), to enhance the smoothness of sequential recommendation models in both data representations and model learning. Specifically, with the help of a complementary model, we smooth and enrich the one-hot representations of contexts and labels to better depict the underlying user preference (i.e., context smoothness and label smoothness), and devise a model regularization strategy to enforce the neighborhood smoothness of the model itself (i.e., model smoothness). Based on these strategies, we design three regularizers to constrain and improve the training of sequential recommendation models. Extensive experiments on five datasets show that our approach is able to improve the performance of various base models consistently and outperform other regularization training methods.

In this paper, we study how to continually pretrain language models for improving the understanding of math problems.Specifically, we focus on solving a fundamental challenge in modeling math problems, i.e., how to fuse the semantics of textual description and formulas, which are highly different in essence.To address this issue, we propose a new approach called COMUS to continually pre-train language models for math problem understanding with syntax-aware memory network.In this approach, we first construct the math syntax graph to model the structural semantic information, by combining the parsing trees of the text and formulas, and then design the syntax-aware memory networks to deeply fuse the features from the graph and text.With the help of syntax relations, we can model the interaction between the token from the text and its semantic-related nodes within the formulas, which is helpful to capture fine-grained semantic correlations between texts and formulas.Besides, we devise three continual pre-training tasks to further align and fuse the representations of the text and math syntax graph.Experimental results on four tasks in the math domain demonstrate the effectiveness of our approach.Our code and data are publicly available at the link: https: //github.com/RUCAIBox/COMUS.

Recently, contrastive learning has been shown to be effective in improving pre-trained language models (PLM) to derive high-quality sentence representations. It aims to pull close positive examples to enhance the alignment while push apart irrelevant negatives for the uniformity of the whole representation space.However, previous works mostly adopt in-batch negatives or sample from training data at random. Such a way may cause the sampling bias that improper negatives (false negatives and anisotropy representations) are used to learn sentence representations, which will hurt the uniformity of the representation space.To address it, we present a new framework DCLR (Debiased Contrastive Learning of unsupervised sentence Representations) to alleviate the influence of these improper negatives.In DCLR, we design an instance weighting method to punish false negatives and generate noise-based negatives to guarantee the uniformity of the representation space.Experiments on seven semantic textual similarity tasks show that our approach is more effective than competitive baselines. Our code and data are publicly available at the link: bluehttps://github.com/RUCAIBox/DCLR.

Within online platforms, it is critical to capture the semantics of sequential user behaviors for accurately predicting user interests. Recently, significant progress has been made in sequential recommendation with deep learning. However, existing neural sequential recommendation models may not perform well in practice due to the sparsity of the real-world data especially in cold-start scenarios. To tackle this problem, we propose the model ReDA, which stands for Retrieval-enhanced Data Augmentation for modeling sequential user behaviors. The main idea of our approach is to leverage the related information from similar users for generating both relevant and diverse augmentation. First, we train a neural retriever to retrieve the augmentation users according to the se- mantic similarity between user representations, and then conduct two types of data augmentation to generate augmented user representations. Furthermore, these augmented data are incorporated in a contrastive learning framework for learning more capable representations. Extensive experiments conducted on both public and industry datasets demonstrate the superiority of our proposed method over existing state-of-the-art methods, especially when only limited training data is available.

In this paper, we study the task of cold-start sequential recommendation, where new users with very short interaction sequences come with time. We cast this problem as a few-shot learning problem and adopt a meta-learning approach to developing our solution. For our task, a major obstacle of effective knowledge transfer that is there exists significant characteristic divergence between old and new interaction sequences for meta-learning. To address the above issues, we purpose a Multimodal MetaLearning (denoted as MML) approach that incorporates multimodal side information of items (e.g., text and image) into the meta-learning process, to stabilize and improve the meta-learning process for cold-start sequential recommendation. In specific, we design a group of multimodal meta-learners corresponding to each kind of modality, where ID features are used to develop the main meta-learner and the rest text and image features are used to develop auxiliary meta-learners. Instead of simply combing the predictions from different meta-learners, we design an adaptive, learnable fusion layer to integrate the predictions based on different modalities. Meanwhile, we design a cold-start item embedding generator, which utilize multimodal side information to warm up the ID embeddings of new items. Extensive offline and online experiments demonstrate that MML can significantly improve the recommendation performance for cold-start users compared with baseline models. Our code is released at https://github.com/RUCAIBox/MML.

The availability of electronic health record data makes it possible to develop automatic disease diagnosis approaches. In this paper, we study the early diagnosis of diseases. As being a difficult task (even for experienced doctors), early diagnosis of diseases poses several challenges that are not well solved by prior studies, including insufficient training data, dynamic and complex signs of complications and trade-off between earliness and accuracy.

Recently, pre-training based approaches are proposed to leverage self-supervised signals for improving the performance of sequential recommendation. However, most of existing pre-training recommender systems simply model the historical behavior of a user as a sequence, while lack of sufficient consideration on temporal interaction patterns that are useful for modeling user behavior.

In order to support the study of recent advances in recommender systems, this paper presents an extended recommendation library consisting of eight packages for up-to-date topics and architectures. First of all, from a data perspective, we consider three important topics related to data issues (ie sparsity, bias and distribution shift ), and develop five packages accordingly, including meta-learning, data augmentation, debiasing, fairness and cross-domain recommendation. Furthermore, from a model perspective, we develop two benchmarking packages for Transformer-based and graph neural network~(GNN)-based models, respectively. All the packages (consisting of 65 new models) are developed based on a popular recommendation framework RecBole, ensuring that both the implementation and interface are unified. For each package, we provide complete implementations from data loading, experimental setup, evaluation and algorithm implementation. This library provides a valuable resource to facilitate the up-to-date research in recommender systems. The project is released at the link: \urlhttps://github.com/RUCAIBox/RecBole2.0.

Recently, Mixture-of-Experts (short as MoE) architecture has achieved remarkable success in increasing the model capacity of large-scale language models. However, MoE requires incorporating significantly more parameters than the base model being extended. In this paper, we propose building a parameter-efficient MoE architecture by sharing information across experts. We adopt matrix product operator (MPO, a tensor decomposition from quantum many-body physics) to reconstruct the parameter matrix in the expert layer and increase model capacity for pre-trained language models by sharing parameters of the central tensor (containing the core information) among different experts while enabling the specificity through the auxiliary tensors (complementing the central tensor) of different experts. To address the unbalanced optimization issue, we further design the gradient mask strategy for the MPO-based MoE architecture. Extensive experiments based on T5 and GPT-2 show improved performance and efficiency of the pre-trained language model (27.2x reduction in total parameters for the superior model performance, compared with the Switch Transformers). Our code is publicly available at https://github.com/RUCAIBox/MPO/MPOE.

Recently, pretrained language models (PLMs) have had exceptional success in language generation. To leverage the rich knowledge encoded by PLMs, a simple yet powerful paradigm is to use prompts in the form of either discrete tokens or continuous embeddings. In existing studies, these prompting methods are typically independent of the inputs, lacking sufficient consideration of input semantics. To address this issue, we propose a novel continuous prompting approach, called context-tuning, to fine-tuning PLMs for natural language generation. Firstly, the prompts are derived based on the input text to elicit useful knowledge from PLMs for generation. We refer to such prompts as contextualized prompts. Secondly, we use continuous inverse prompting to improve the process of natural language generation by modeling an inverse generation process from output to input, making the generated text more relevant to the inputs. Furthermore, we utilize a lightweight context-tuning method that fine-tunes only 0.12% of the parameters while maintaining good performance. Our code is publicly available at https://github.com/RUCAIBox/Context-Tuning.

We study the text generation task under the approach of pre-trained language models (PLMs).Typically, an auto-regressive (AR) method is adopted for generating texts in a token-bytoken manner.Despite many advantages of AR generation, it usually suffers from inefficient inference.Therefore, non-autoregressive (NAR) models are proposed to generate all target tokens simultaneously.However, NAR models usually generate texts of lower quality due to the absence of token dependency in the output text.In this paper, we propose ELMER: an Efficient and effective PLM for NAR tExt geneRation to explicitly model the token dependency during NAR generation.By leveraging the early exit technique, ELMER enables the token generations at different layers, according to their prediction confidence (a more confident token will exit at a lower layer).Besides, we propose a novel pre-training objective, Layer Permutation Language Modeling, to pre-train ELMER by permuting the exit layer for each token in sequences.Experiments on three text generation tasks show that ELMER significantly outperforms NAR models and further narrows the performance gap with AR PLMs (e.g., ELMER (29.92) vs BART (30.61)ROUGE-L in XSUM) while achieving over 10 times inference speedup.

To develop effective and efficient graph similarity learning (GSL) models, a series of data-driven neural algorithms have been proposed in recent years. Although GSL models are frequently deployed in privacy-sensitive scenarios, the user privacy protection of neural GSL models has not drawn much attention. To comprehensively understand the privacy protection issues, we first introduce the concept of attackable representation to systematically characterize the privacy attacks that each model can face. Inspired by the qualitative results, we propose a novel P rivacy-P reserving neural G raph M atching network model, named PPGM, for graph similarity learning. To prevent reconstruction attacks, the proposed model does not communicate node-level representations between devices. Instead, we learn multi-perspective graph representations based on learnable context vectors. To alleviate the attacks to graph properties, the obfuscated features that contain information from both graphs are communicated. In this way, the private properties of each graph can be difficult to infer. Based on the node-graph matching techniques while calculating the obfuscated features, PPGM can also be effective in similarity measuring. To quantitatively evaluate the privacy-preserving ability of neural GSL models, we further propose an evaluation protocol via training supervised black-box attack models. Extensive experiments on widely-used benchmarks show the effectiveness and strong privacy-protection ability of the proposed model PPGM. The code is available at: https://github.com/RUCAIBox/PPGM.

As transformer evolves, pre-trained models have advanced at a breakneck pace in recent years. They have dominated the mainstream techniques in natural language processing (NLP) and computer vision (CV). How to adapt pre-training to the field of Vision-and-Language (V-L) learning and improve downstream task performance becomes a focus of multimodal learning. In this paper, we review the recent progress in Vision-Language Pre-Trained Models (VL-PTMs). As the core content, we first briefly introduce several ways to encode raw images and texts to single-modal embeddings before pre-training. Then, we dive into the mainstream architectures of VL-PTMs in modeling the interaction between text and image representations. We further present widely-used pre-training tasks, and then we introduce some common downstream tasks. We finally conclude this paper and present some promising research directions. Our survey aims to provide researchers with synthesis and pointer to related research.

The Question Answering over Knowledge Graph (KGQA) task seeks entities (answers) from the Knowledge Graph (KG) in order to answer natural language questions. In practice, KG is often incomplete, with numerous missing links and nodes. With such an incomplete KG, it is tricky to use the semantics inside the KG to get the golden answers, particularly for complex questions. Some current efforts concentrate on using external corpora to overcome KG sparsity; however, identifying and obtaining the corpora is challenging. Other types of work aim to leverage the pre-trained embeddings to resolve the issue but perform slightly worse on complex questions involving numerous triple facts in KG. To address the aforementioned problems, we present a framework CAPKGQA, which transforms Complex KGQA into an n-Ary link Prediction task capable of explicitly modeling complex questions. Furthermore, previous methods also suffer from incomplete KG throughout the candidate answer generation phase. Therefore, we devise an embedding-based retrieval strategy to extract more reliable candidate answers from incomplete KG. Extensive experiments reveal that our approach beats the state-of-the-art models on incomplete and complex KGQA tasks by a significant margin.

In order to develop effective sequential recommenders, a series of sequence representation learning (SRL) methods are proposed to model historical user behaviors. Most existing SRL methods rely on explicit item IDs for developing the sequence models to better capture user preference. Though effective to some extent, these methods are difficult to be transferred to new recommendation scenarios, due to the limitation by explicitly modeling item IDs. To tackle this issue, we present a novel universal sequence representation learning approach, named UniSRec. The proposed approach utilizes the associated description text of items to learn transferable representations across different recommendation scenarios. For learning universal item representations, we design a lightweight item encoding architecture based on parametric whitening and mixture-of-experts enhanced adaptor. For learning universal sequence representations, we introduce two contrastive pre-training tasks by sampling multi-domain negatives. With the pre-trained universal sequence representation model, our approach can be effectively transferred to new recommendation domains or platforms in a parameter-efficient way, under either inductive or transductive settings. Extensive experiments conducted on real-world datasets demonstrate the effectiveness of the proposed approach. Especially, our approach also leads to a performance improvement in a cross-platform setting, showing the strong transferability of the proposed universal SRL method. The code and pre-trained model are available at: https://github.com/RUCAIBox/UniSRec.

Relevant recommendation is a special recommendation scenario which provides relevant items when users express interests on one target item (e.g., click, like and purchase). Besides considering the relevance between recommendations and trigger item, the recommendations should also be diversified to avoid information cocoons. However, existing diversified recommendation methods mainly focus on item-level diversity which is insufficient when the recommended items are all relevant to the target item. Moreover, redundant or noisy item features might affect the performance of simple feature-aware recommendation approaches. Faced with these issues, we propose a Feature Disentanglement Self-Balancing Re-ranking framework (FDSB) to capture feature- aware diversity. The framework consists of two major modules, namely disentangled attention encoder (DAE) and self-balanced multi-aspect ranker. In DAE, we use multi-head attention to learn disentangled aspects from rich item features. In the ranker, we develop an aspect-specific ranking mechanism that is able to adaptively balance the relevance and diversity for each aspect. In experiments, we conduct offline evaluation on the collected dataset and deploy FDSB on KuaiShou app for online ??/?? test on the function of relevant recommendation. The significant improvements on both recommendation quality and user experience verify the effectiveness of our approach.

In recent years, automatic computational systems based on deep learning are widely used in medical fields, such as automatic diagnosing and disease prediction. Most of these systems are designed for data sufficient scenarios. However, due to the disease rarity or privacy, the medical data are always insufficient. When applying these data-hungry deep learning models with insufficient data, it is likely to lead to issues of over-fitting and cause serious performance problems. Many data augmentation methods have been proposed to solve the data insufficiency problem, such as using GAN (Generative Adversarial Networks) to generate training data. However, the augmented data usually contains lots of noise. Directly using them to train sensitive medical models is very difficult to achieve satisfactory results.

Conversational recommender systems (CRS) aim to proactively elicit user preference and recommend high-quality items through natural language conversations. Typically, a CRS consists of a recommendation module to predict preferred items for users and a conversation module to generate appropriate responses. To develop an effective CRS, it is essential to seamlessly integrate the two modules. Existing works either design semantic alignment strategies, or share knowledge resources and representations between the two modules. However, these approaches still rely on different architectures or techniques to develop the two modules, making it difficult for effective module integration. To address this problem, we propose a unified CRS model named UniCRS based on knowledge-enhanced prompt learning. Our approach unifies the recommendation and conversation subtasks into the prompt learning paradigm, and utilizes knowledge-enhanced prompts based on a fixed pre-trained language model (PLM) to fulfill both subtasks in a unified approach. In the prompt design, we include fused knowledge representations, task-specific soft tokens, and the dialogue context, which can provide sufficient contextual information to adapt the PLM for the CRS task. Besides, for the recommendation subtask, we also incorporate the generated response template as an important part of the prompt, to enhance the information interaction between the two subtasks. Extensive experiments on two public CRS datasets have demonstrated the effectiveness of our approach. Our code is publicly available at the link: https://github.com/RUCAIBox/UniCRS.

This paper aims to advance the mathematical intelligence of machines by presenting the first Chinese mathematical pre-trained language model (PLM) for effectively understanding and representing mathematical problems. Unlike other standard NLP tasks, mathematical texts are difficult to understand, since they involve mathematical terminology, symbols and formulas in the problem statement. Typically, it requires complex mathematical logic and background knowledge for solving mathematical problems.

Junyi Li, Tianyi Tang, Zheng Gong, Lixin Yang, Zhuohao Yu, Zhipeng Chen, Jingyuan Wang, Xin Zhao, Ji-Rong Wen. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Junyi Li, Tianyi Tang, Jian-Yun Nie, Ji-Rong Wen, Xin Zhao. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Person-job fit is the core technique of online recruitment platforms, which can improve the efficiency of recruitment by accurately matching the job positions with the job seekers. Existing works mainly focus on modeling the unidirectional process or overall matching. However, recruitment is a two-way selection process, which means that both candidate and employer involved in the interaction should meet the expectation of each other, instead of unilateral satisfaction. In this paper, we propose a dual-perspective graph representation learning approach to model directed interactions between candidates and jobs. To model the two-way selection preference from the dual-perspective of job seekers and employers, we incorporate two different nodes for each candidate (or job) and characterize both successful matching and failed matching via a unified dual-perspective interaction graph. To learn dual-perspective node representations effectively, we design an effective optimization algorithm, which involves a quadruple-based loss and a dual-perspective contrastive learning loss. Extensive experiments on three large real-world recruitment datasets have shown the effectiveness of our approach. Our code is available at https://github.com/RUCAIBox/DPGNN .

Recently, graph neural networks (GNNs) have been shown powerful capacity at modeling structural data. However, when adapted to downstream tasks, it usually requires abundant task-specific labeled data, which can be extremely scarce in practice. A promising solution to data scarcity is to pre-train a transferable and expressive GNN model on large amounts of unlabeled graphs or coarse-grained labeled graphs. Then the pre-trained GNN is fine-tuned on downstream datasets with task-specific fine-grained labels. In this paper, we present a novel Graph Matching based GNN Pre-Training framework, called GMPT. Focusing on a pair of graphs, we propose to learn structural correspondences between them via neural graph matching, consisting of both intra-graph message passing and inter-graph message passing. In this way, we can learn adaptive representations for a given graph when paired with different graphs, and both node- and graph-level characteristics are naturally considered in a single pre-training task. The proposed method can be applied to fully self-supervised pre-training and coarse-grained supervised pre-training. We further propose an approximate contrastive training strategy to significantly reduce time/memory consumption. Extensive experiments on multi-domain, out-of-distribution benchmarks have demonstrated the effectiveness of our approach. The code is available at: https://github.com/RUCAIBox/GMPT.

A large-scale recommender system usually consists of recall and ranking modules. The goal of ranking modules (aka rankers) is to elaborately discriminate users' preference on item candidates proposed by recall modules. With the success of deep learning techniques in various domains, we have witnessed the mainstream rankers evolve from traditional models to deep neural models. However, the way that we design and use rankers remains unchanged: offline training the model, freezing the parameters, and deploying it for online serving. Actually, the candidate items are determined by specific user requests, in which underlying distributions (e.g., the proportion of items for different categories, the proportion of popular or new items) are highly different from one another in a production environment. The classical parameter-frozen inference manner cannot adapt to dynamic serving circumstances, making rankers' performance compromised.

Session-based Recommendation (SBR) refers to the task of predicting the next item based on short-term user behaviors within an anonymous session. However, session embedding learned by a non-linear encoder is usually not in the same representation space as item embeddings, resulting in the inconsistent prediction issue while recommending items. To address this issue, we propose a simple and effective framework named CORE, which can unify the representation space for both the encoding and decoding processes. Firstly, we design a representation-consistent encoder that takes the linear combination of input item embeddings as session embedding, guaranteeing that sessions and items are in the same representation space. Besides, we propose a robust distance measuring method to prevent overfitting of embeddings in the consistent representation space. Extensive experiments conducted on five public real-world datasets demonstrate the effectiveness and efficiency of the proposed method. The code is available at: https://github.com/RUCAIBox/CORE.

Limited by the statistical-based machine learning framework, a spurious correlation is likely to appear in existing knowledge-aware recommendation methods. It refers to a knowledge fact that appears causal to the user behaviors (inferred by the recommender) but is not in fact. For tackling this issue, we present a novel approach to discovering and alleviating the potential spurious correlations from a counterfactual perspective. To be specific, our approach consists of two counterfactual generators and a recommender. The counterfactual generators are designed to generate counterfactual interactions via reinforcement learning, while the recommender is implemented with two different graph neural networks to aggregate the information from KG and user-item interactions respectively. The counterfactual generators and recommender are integrated in a mutually collaborative way. With this approach, the recommender helps the counterfactual generators better identify potential spurious correlations and generate high-quality counterfactual interactions, while the counterfactual generators help the recommender weaken the influence of the potential spurious correlations simultaneously. Extensive experiments on three real-world datasets have shown the effectiveness of the proposed approach by comparing it with a number of competitive baselines. Our implementation code is available at: https://github.com/RUCAIBox/CGKR.

Recently, graph neural networks (GNN) have been successfully applied to recommender systems as an effective collaborative filtering (CF) approach. However, existing GNN-based CF models suffer from noisy user-item interaction data, which seriously affects the effectiveness and robustness in real-world applications. Although there have been several studies on data denoising in recommender systems, they either neglect direct intervention of noisy interaction in the message-propagation of GNN, or fail to preserve the diversity of recommendation when denoising.

Conversational recommender systems (CRS) aim to recommend suitable items to users through natural language conversations. For developing effective CRSs, a major technical issue is how to accurately infer user preference from very limited conversation context. To address issue, a promising solution is to incorporate external data for enriching the context information. However, prior studies mainly focus on designing fusion models tailored for some specific type of external data, which is not general to model and utilize multi-type external data. To effectively leverage multi-type external data, we propose a novel coarse-to-fine contrastive learning framework to improve data semantic fusion for CRS. In our approach, we first extract and represent multi-grained semantic units from different data signals, and then align the associated multi-type semantic units in a coarse-to-fine way. To implement this framework, we design both coarse-grained and fine-grained procedures for modeling user preference, where the former focuses on more general, coarse-grained semantic fusion and the latter focuses on more specific, fine-grained semantic fusion. Such an approach can be extended to incorporate more kinds of external data. Extensive experiments on two public CRS datasets have demonstrated the effectiveness of our approach in both recommendation and conversation tasks.

Recently, graph collaborative filtering methods have been proposed as an effective recommendation approach, which can capture users’ preference over items by modeling the user-item interaction graphs. Despite the effectiveness, these methods suffer from data sparsity in real scenarios. In order to reduce the influence of data sparsity, contrastive learning is adopted in graph collaborative filtering for enhancing the performance. However, these methods typically construct the contrastive pairs by random sampling, which neglect the neighboring relations among users (or items) and fail to fully exploit the potential of contrastive learning for recommendation.

Recently, deep neural networks such as RNN, CNN and Transformer have been applied in the task of sequential recommendation, which aims to capture the dynamic preference characteristics from logged user behavior data for accurate recommendation. However, in online platforms, logged user behavior data is inevitable to contain noise, and deep recommendation models are easy to overfit on these logged data. To tackle this problem, we borrow the idea of filtering algorithms from signal processing that attenuates the noise in the frequency domain. In our empirical experiments, we find that filtering algorithms can substantially improve representative sequential recommendation models, and integrating simple filtering algorithms (e.g., Band-Stop Filter) with an all-MLP architecture can even outperform competitive Transformer-based models. Motivated by it, we propose FMLP-Rec, an all-MLP model with learnable filters for sequential recommendation task. The all-MLP architecture endows our model with lower time complexity, and the learnable filters can adaptively attenuate the noise information in the frequency domain. Extensive experiments conducted on eight real-world datasets demonstrate the superiority of our proposed method over competitive RNN, CNN, GNN and Transformer-based methods. Our code and data are publicly available at the link: https://github.com/RUCAIBox/FMLP-Rec .

In this work, we study an important task in location-based services, namelyPersonalized Route Recommendation (PRR). Given a road network, the PRR task aims to generate user-specific route suggestions for replying to users’ route queries. A classic approach is to adapt search algorithms to construct pathfinding-like solutions. These methods typically focus on reducing search space with suitable heuristic strategies. For these search algorithms, heuristic strategies are often handcrafted, which are not flexible to work in complicated task settings. In addition, it is difficult to utilize useful context information in the search procedure. To develop a more principled solution to the PRR task, we propose to improve search algorithms with neural networks for solving the PRR task based on the widely used$A^{*}$algorithm. The main idea of our solution is to automatically learn the cost functions in$A^{*}$algorithms, which is the key of heuristic search algorithms. Our model consists of two main components. First, we employ attention-based Recurrent Neural Networks (RNN) to model the cost from the source to the candidate location by incorporating useful context information. Instead of learning a single cost value, the RNN component is able to learn a time-varying vectorized representation for the moving state of a user. Second, we propose to use an estimation network for predicting the cost from a candidate location to the destination. For capturing structural characteristics, the estimation network is built on top of position-aware graph attention networks. The two components are integrated in a principled way for deriving a more accurate cost of a candidate location for the$A^{*}$algorithm. Extensive experiment results on three real-world datasets have shown the effectiveness and robustness of the proposed model.

Citation count prediction is an important task for estimating the future impact of research papers. Most of the existing works utilize the information extracted from the paper itself. In this article, we focus on how to utilize another kind of useful data signal (i.e., peer review text) to improve both the performance and interpretability of the prediction models. Specially, we propose a novel aspect-aware capsule network for citation count prediction based on review text. It contains two major capsule layers, namely the feature capsule layer and the aspect capsule layer, with two different routing approaches, respectively. Feature capsules encode the local semantics from review sentences as the input of aspect capsule layer, whereas aspect capsules aim to capture high-level semantic features that will be served as final representations for prediction. Besides the predictive capacity, we also enhance the model interpretability with two strategies. First, we use the topic distribution of the review text to guide the learning of aspect capsules so that each aspect capsule can represent a specific aspect in the review. Then, we use the learned aspect capsules to generate readable text for explaining the predicted citation count. Extensive experiments on two real-world datasets have demonstrated the effectiveness of the proposed model in both performance and interpretability.

In recommender systems, it is essential to understand the underlying factors that affect user-item interaction. Recently, several studies have utilized disentangled representation learning to discover such hidden factors from user-item interaction data, which shows promising results. However, without any external guidance signal, the learned disentangled representations lack clear meanings, and are easy to suffer from the data sparsity issue. In light of these challenges, we study how to leverage knowledge graph (KG) to guide the disentangled representation learning in recommender systems. The purpose for incorporating KG is twofold, making the disentangled representations interpretable and resolving data sparsity issue. However, it is not straightforward to incorporate KG for improving disentangled representations, because KG has very different data characteristics compared with user-item interactions. We propose a novel K nowledge-guided D isentangled R epresentations approach ( KDR ) to utilizing KG to guide the disentangled representation learning in recommender systems. The basic idea, is to first learn more interpretable disentangled dimensions (explicit disentangled representations) based on structural KG, and then align implicit disentangled representations learned from user-item interaction with the explicit disentangled representations. We design a novel alignment strategy based on mutual information maximization. It enables the KG information to guide the implicit disentangled representation learning, and such learned disentangled representations will correspond to semantic information derived from KG. Finally, the fused disentangled representations are optimized to improve the recommendation performance. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed model in terms of both performance and interpretability.

Session-based Recommendations recommend the next possible item for the user with anonymous sessions, whose challenge is that the user’s behavioral preference can only be analyzed in a limited sequence to meet their need. Recent advances evaluate the effectiveness of the attention mechanism in the session-based recommendation. However, two simplifying assumptions are made by most of these attention-based models. One is to regard the last-click as the query vector to denote the user’s current preference, and the other is to consider that all items within the session are favorable for the final result, including the effect of unrelated items (i.e., spurious user behaviors). In this paper, we propose a novel Dual Sparse Attention Network for the session-based recommendation called DSAN to address these shortcomings. In this proposed method, we explore a learned target item embedding to model the user’s current preference and apply an adaptively sparse transformation function to eliminate the effect of the unrelated items. Experimental results on two real public datasets show that the proposed method is superior to the state-of-the-art session-based recommendation algorithm in all tests and also demonstrate that not all actions within the session are useful. To make our results reproducible, we have published our code on https://github.com/SamHaoYuan/DSANForAAAI2021.

Peiyu Liu, Ze-Feng Gao, Wayne Xin Zhao, Zhi-Yuan Xie, Zhong-Yi Lu, Ji-Rong Wen. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Within online platforms, it is critical to capture the semantics of sequential user behaviors for accurately modeling user interests. However, dynamic characteristics and sparse behaviors make it difficult to train effective user representations for sequential user behavior modeling.

Incorporating review information into the recommender system has been demonstrated to be an effective method for boosting the recommendation performance. Previous research mainly focus on designing advanced architectures to better profile the users and items. However, the review information in realities can be highly sparse and imbalanced, which poses great challenges for effective user/item representations and satisfied performance enhancement. To alleviate this problem, in this paper, we propose to improve review-based recommendation by counterfactually augmenting the training samples. We focus on a common setting --- feature-aware recommendation, and the main building block of our idea lies in the counterfactual question: "what would be the user's decision if her feature-level preference had been different?''. When augmenting the training samples, we actively change the user preference (also called intervention), and predict the user feedback on the items based on pre-trained recommender models. Instead of changing the user preference in a random manner, we design a learning-based method to discover the samples which are more effective for model optimization. In order to improve the sample qualities, we propose two strategies --- constrained feature perturbation and frequency-based sampling --- to equip our model. Since the sample generation model can be not perfect, we theoretically analyze the relation between the model prediction error and the number of generated samples. As a byproduct, our framework can explain the user pair-wise preference, which is complementary to the traditional point-wise explanations. Extensive experiments demonstrate that our model can significantly improve the performance of the state-of-the-art methods.

Attention-based recommender models hold the promise of improving performance by learning to discriminate different user/item feature importances. However, due to the existence of the latent confounders, the correlations captured by attention mechanisms may fail to reflect the true influence of the features on the targets (i.e., spurious correlation). In this paper, we propose to empower attention mechanism by the causal inference, which is a powerful tool to identify the real causal effects. Our model is based on the potential outcome framework, where the item features are regarded as the treatment and the outcome is the predicted user preference. In specific, the causal relation of each feature on the outcome is measured by the individual treatment effect (ITE). In order to distill the causal information into the attention learning process, we minimize the distance between the traditional attention weights and the normalized ITE. With such causal regularization, the learned attention weights can capture the real causal effects, which are expected to correct the feature importances for improving performance. We conduct extensive experiments based on three real-world datasets to demonstrate the effectiveness.

In recent years, there are a large number of recommendation algorithms proposed in the literature, from traditional collaborative filtering to deep learning algorithms. However, the concerns about how to standardize open source implementation of recommendation algorithms continually increase in the research community. In the light of this challenge, we propose a unified, comprehensive and efficient recommender system library called RecBole (pronounced as [rEk'[email protected]]), which provides a unified framework to develop and reproduce recommendation algorithms for research purpose. In this library, we implement 73 recommendation models on 28 benchmark datasets, covering the categories of general recommendation, sequential recommendation, context-aware recommendation and knowledge-based recommendation. We implement the RecBole library based on PyTorch, which is one of the most popular deep learning frameworks. Our library is featured in many aspects, including general and extensible data structures, comprehensive benchmark models and datasets, efficient GPU-accelerated execution, and extensive and standard evaluation protocols. We provide a series of auxiliary functions, tools, and scripts to facilitate the use of this library, such as automatic parameter tuning and break-point resume. Such a framework is useful to standardize the implementation and evaluation of recommender systems. The project and documents are released at https://recbole.io/.

In various natural language processing tasks, passage retrieval and passage re-ranking are two key procedures in finding and ranking relevant information.Since both the two procedures contribute to the final performance, it is important to jointly optimize them in order to achieve mutual improvement.In this paper, we propose a novel joint training approach for dense passage retrieval and passage reranking.A major contribution is that we introduce the dynamic listwise distillation, where we design a unified listwise training approach for both the retriever and the re-ranker.During the dynamic distillation, the retriever and the re-ranker can be adaptively improved according to each other's relevance information.We also propose a hybrid data augmentation strategy to construct diverse training instances for listwise training approach.Extensive experiments show the effectiveness of our approach on both MSMARCO and Natural Questions datasets.Our code is available at https:// github.com/PaddlePaddle/RocketQA.

Recent works have shown that powerful pretrained language models (PLM) can be fooled by small perturbations or intentional attacks.To solve this issue, various data augmentation techniques are proposed to improve the robustness of PLMs.However, it is still challenging to augment semantically relevant examples with sufficient diversity.In this work, we present Virtual Data Augmentation (VDA), a general framework for robustly fine-tuning PLMs.Based on the original token embeddings, we construct a multinomial mixture for augmenting virtual data embeddings, where a masked language model guarantees the semantic relevance and the Gaussian noise provides the augmentation diversity.Furthermore, a regularized training strategy is proposed to balance the two aspects.Extensive experiments on six datasets show that our approach is able to improve the robustness of PLMs and alleviate the performance degradation under adversarial attacks.

With the increase of traffic prediction models, there has become an urgent need to develop a standardized framework to implement and evaluate these methods. This paper presents LibCity, a unified, comprehensive, and extensible library for traffic prediction, which provides researchers with a credible experimental tool and a convenient development framework. In this library, we reproduce 42 traffic prediction models and collect 29 spatial-temporal datasets, which allows researchers to conduct comprehensive experiments in a convenient way. To accelerate the development of new models, we design unified model interfaces based on unified data formats, which effectively encapsulate the details of the implementation. To verify the effectiveness of our implementations, we also report the reproducibility comparison results of LibCity, and set up a performance leaderboard for the four kinds of traffic prediction tasks. Our library will contribute to the standardization and reproducibility in the field of traffic prediction. The open source link of LibCity is https://github.com/LibCity/Bigscity-LibCity.

Knowledge base question answering (KBQA) aims to answer a question over a knowledge base (KB). Recently, a large number of studies focus on semantically or syntactically complicated questions. In this paper, we elaborately summarize the typical challenges and solutions for complex KBQA. We begin with introducing the background about the KBQA task. Next, we present the two mainstream categories of methods for complex KBQA, namely semantic parsing-based (SP-based) methods and information retrieval-based (IR-based) methods. We then review the advanced methods comprehensively from the perspective of the two categories. Specifically, we explicate their solutions to the typical challenges. Finally, we conclude and discuss some promising directions for future research.

Text generation has become one of the most important yet challenging tasks in natural language processing (NLP). The resurgence of deep learning has greatly advanced this field by neural generation models, especially the paradigm of pretrained language models (PLMs). In this paper, we present an overview of the major advances achieved in the topic of PLMs for text generation. As the preliminaries, we present the general task definition and briefly describe the mainstream architectures of PLMs for text generation. As the core content, we discuss how to adapt existing PLMs to model different input data and satisfy special properties in the generated text. We further summarize several important fine-tuning strategies for text generation. Finally, we present several future directions and conclude this paper. Our survey aims to provide text generation researchers a synthesis and pointer to related research.

As it becomes prevalent that user information exists in multiple platforms or services, cross-domain recommendation has been an important task in industry. Although it is well known that users tend to show different preferences in different domains, existing studies seldom model how domain biases affect user preferences. Focused on this issue, we develop a casual-based approach to mitigating the domain biases when transferring the user information cross domains. To be specific, this paper presents a novel debiasing learning based cross-domain recommendation framework with causal embedding. In this framework, we design a novel Inverse-Propensity-Score (IPS) estimator designed for cross-domain scenario, and further propose three kinds of restrictions for propensity score learning. Our framework can be generally applied to various recommendation algorithms for cross-domain recommendation. Extensive experiments on both public and industry datasets have demonstrated the effectiveness of the proposed framework.

With the development of electronic health records (EHRs), prenatal care examination records have become available for developing automatic prediction or diagnosis approaches with machine learning methods. In this paper, we study how to effectively learn representations applied to various downstream tasks for EHR data. Although several methods have been proposed in this direction, they usually adapt classic sequential models to solve one specific diagnosis task or address unique EHR data issues. This makes it difficult to reuse these existing methods for the early diagnosis of pregnancy complications or provide a general solution to address the series of health problems caused by pregnancy complications. In this paper, we propose a novel model RAPT, which stands for RepresentAtion by Pre-training time-aware Transformer. To associate pre-training and EHR data, we design an architecture that is suitable for both modeling EHR data and pre-training, namely time-aware Transformer. To handle various characteristics in EHR data, such as insufficiency, we carefully devise three pre-training tasks to handle data insufficiency, data incompleteness and short sequence problems, namely similarity prediction, masked prediction and reasonability check. In this way, our representations can capture various EHR data characteristics. Extensive experimental results for four downstream tasks have shown the effectiveness of the proposed approach. We also introduce sensitivity analysis to interpret the model and design an interface to show results and interpretation for doctors. Finally, we implement a diagnosis system for pregnancy complications based on our pre-training model. Doctors and pregnant women can benefit from the diagnosis system in early diagnosis of pregnancy complications.

In industry, customer lifetime value (LTV) prediction is a challenging task, since user consumption data is usually volatile, noisy, or sparse. To address these issues, this paper presents a novel Temporal-Structural User Representation (named TSUR) network to predict LTV. We utilize historical revenue time series and user attributes to learn both temporal and structural user representations, respectively. Specifically, the temporal representation is learned with a temporal trend encoder based on a novel multi-channel Discrete Wavelet Transform~(DWT) module, while the structural representation is derived with Graph Attention Network (GAT) on an attribute similarity graph. Furthermore, a novel cluster-alignment regularization method is employed to align and enhance these two kinds of representations. In essence, such a fusion way can be considered as the association of temporal and structural representations in the low-pass representation space, which is also useful to prevent the data noise from being transferred across different views. To our knowledge, it is the first time that temporal and structural user representations are jointly learned for LTV prediction. Extensive offline experiments on two large-scale real-world datasets and online A/B tests have shown the superiority of our approach over a number of competitive baselines.

Recent studies reveal that recommender systems are vulnerable to data poisoning attack due to their openness nature. In data poisoning attack, the attacker typically recruits a group of controlled users to inject well-crafted user-item interaction data into the recommendation model's training set to modify the model parameters as desired. Thus, existing attack approaches usually require full access to the training data to infer items' characteristics and craft the fake interactions for controlled users. However, such attack approaches may not be feasible in practice due to the attacker's limited data collection capability and the restricted access to the training data, which sometimes are even perturbed by the privacy preserving mechanism of the service providers. Such design-reality gap may cause failure of attacks. In this paper, we fill the gap by proposing two novel adversarial attack approaches to handle the incompleteness and perturbations in user-item interaction data. First, we propose a bi-level optimization framework that incorporates a probabilistic generative model to find the users and items whose interaction data is sufficient and has not been significantly perturbed, and leverage these users and items' data to craft fake user-item interactions. Moreover, we reverse the learning process of recommendation models and develop a simple yet effective approach that can incorporate context-specific heuristic rules to handle data incompleteness and perturbations. Extensive experiments on two datasets against three representative recommendation models show that the proposed approaches can achieve better attack performance than existing approaches.

Yingqi Qu, Yuchen Ding, Jing Liu, Kai Liu, Ruiyang Ren, Wayne Xin Zhao, Daxiang Dong, Hua Wu, Haifeng Wang. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Self-attention networks (SANs) have been intensively applied for sequential recommenders, but they are limited due to: (1) the quadratic complexity and vulnerability to over-parameterization in self-attention; (2) inaccurate modeling of sequential relations between items due to the implicit position encoding. In this work, we propose the low-rank decomposed self-attention networks (LightSANs) to overcome these problems. Particularly, we introduce the low-rank decomposed self-attention, which projects user's historical items into a small constant number of latent interests and leverages item-to-interest interaction to generate the context-aware representation. It scales linearly w.r.t. the user's historical sequence length in terms of time and space, and is more resilient to over-parameterization. Besides, we design the decoupled position encoding, which models the sequential relations between items more precisely. Extensive experimental studies are carried out on three real-world datasets, where LightSANs outperform the existing SANs-based recommenders in terms of both effectiveness and efficiency.

As a natural language generation task, it is challenging to generate informative and coherent review text. In order to enhance the informativeness of the generated text, existing solutions typically learn to copy entities or triples from knowledge graphs (KGs). However, they lack overall consideration to select and arrange the incorporated knowledge, which tends to cause text incoherence. To address the above issue, we focus on improving entity-centric coherence of the generated reviews by leveraging the semantic structure of KGs. In this paper, we propose a novel Coherence Enhanced Text Planning model (CETP) based on knowledge graphs (KGs) to improve both global and local coherence for review generation. The proposed model learns a two-level text plan for generating a document: (1) the document plan is modeled as a sequence of sentence plans in order, and (2) the sentence plan is modeled as an entity-based subgraph from KG. Local coherence can be naturally enforced by KG subgraphs through intra-sentence correlations between entities. For global coherence, we design a hierarchical self-attentive architecture with both subgraph- and node-level attention to enhance the correlations between subgraphs. To our knowledge, we are the first to utilize a KG-based text planning model to enhance text coherence for review generation. Extensive experiments on three datasets confirm the effectiveness of our model on improving the content coherence of generated texts.

Sequential recommendation aims at predicting users' preferences based on their historical behaviors. However, this recommendation strategy may not perform well in practice due to the sparsity of the real-world data. In this paper, we propose a novel counterfactual data augmentation framework to mitigate the impact of the imperfect training data and empower sequential recommendation models. Our framework is composed of a sampler model and an anchor model. The sampler model aims to generate new user behavior sequences based on the observed ones, while the anchor model is leveraged to provide the final recommendation list, which is trained based on both observed and generated sequences. We design the sampler model to answer the key counterfactual question: "what would a user like to buy if her previously purchased items had been different?". Beyond heuristic intervention methods, we leverage two learning-based methods to implement the sampler model, and thus, improve the quality of the generated sequences when training the anchor model. Additionally, we analyze the influence of the generated sequences on the anchor model in theory and achieve a trade-off between the information and the noise introduced by the generated sequences. Experiments on nine real-world datasets demonstrate our framework's effectiveness and generality.

Multi-hop Knowledge Base Question Answering (KBQA) aims to find the answer entities that are multiple hops away in the Knowl- edge Base (KB) from the entities in the question. A major challenge is the lack of supervision signals at intermediate steps. Therefore, multi-hop KBQA algorithms can only receive the feedback from the final answer, which makes the learning unstable or ineffective. To address this challenge, we propose a novel teacher-student approach for the multi-hop KBQA task. In our approach, the stu- dent network aims to find the correct answer to the query, while the teacher network tries to learn intermediate supervision signals for improving the reasoning capacity of the student network. The major novelty lies in the design of the teacher network, where we utilize both forward and backward reasoning to enhance the learning of intermediate entity distributions. By considering bidi- rectional reasoning, the teacher network can produce more reliable intermediate supervision signals, which can alleviate the issue of spurious reasoning. Extensive experiments on three benchmark datasets have demonstrated the effectiveness of our approach on the KBQA task.

The evolution of mobile apps has greatly changed the way that we live. It becomes increasingly important to understand and model the users on mobile apps. Instead of focusing on some specific app alone, it has become a popular paradigm to study the user behavior on various mobile apps in a symbiotic environment.

With the development of location-acquisition technologies, there are a huge number of mobile trajectories generated and accumulated in a variety of domains. However, due to the constraints of device and environment, many trajectories are recorded at low sampling rate, which increases the uncertainty between two consecutive sampled points in the trajectories. Our task is to recover a high-sampled trajectory based on the irregular low-sampled trajectory in free space, i.e., without road network information. There are two major problems with traditional solutions. First, many of these methods rely on heuristic search algorithms or simple probabilistic models. They cannot well capture complex sequential dependencies or global data correlations. Second, for reducing the predictive complexity of the unconstrained numerical coordinates, most of the previous studies have adopted a common preprocessing strategy by mapping the space into discrete units. As a side effect, using discrete units is likely to bring noise or inaccurate information. Hence, a principled post-calibration step is required to produce accurate results, which has been seldom studied by existing methods. To address the above difficulties, we propose a novel Deep Hybrid Trajectory Recovery model, named DHTR. Our recovery model extends the classic sequence-to-sequence generation framework by implementing a subsequence-to-sequence recovery model tailored for the current task, named subseq2seq. In order to effectively capture spatiotemporal correlations, we adopt both spatial and temporal attentions for enhancing the model performance. With the attention mechanisms, our model is able to characterize long-range correlations among trajectory points. Furthermore, we integrate the subseq2seq with a calibration component of Kalman filter (KF) for reducing the predictive uncertainty. At each timestep, the noisy predictions from the subseq2seq component will be fed into the KF component for calibration, and then the refined predictions will be forwarded to the subseq2seq component for the computation of the next timestep. Extensive results on real-world datasets have shown the superiority of the proposed model in both performance and interpretability.

With the ever-increasing growth of online recruitment data, job-resume matching has become an important task to automatically match jobs with suitable resumes. This task is typically casted as a supervised text matching problem. Supervised learning is powerful when the labeled data is sufficient. However, on online recruitment platforms, job-resume interaction data is sparse and noisy, which affects the performance of job-resume match algorithms.

Personalized review generation (PRG) aims to automatically produce review text reflecting user preference, which is a challenging natural language generation task. Most of previous studies do not explicitly model factual description of products, tending to generate uninformative content. Moreover, they mainly focus on word-level generation, but cannot accurately reflect more abstractive user preference in multiple aspects. To address the above issues, we propose a novel knowledgeenhanced PRG model based on capsule graph neural network (CapsGNN). We first construct a heterogeneous knowledge graph (HKG) for utilizing rich item attributes. We adopt Caps-GNN to learn graph capsules for encoding underlying characteristics from the HKG. Our generation process contains two major steps, namely aspect sequence generation and sentence generation. First, based on graph capsules, we adaptively learn aspect capsules for inferring the aspect sequence. Then, conditioned on the inferred aspect label, we design a graph-based copy mechanism to generate sentences by incorporating related entities or words from HKG. To our knowledge, we are the first to utilize knowledge graph for the PRG task. The incorporated KG information is able to enhance user preference at both aspect and word levels. Extensive experiments on three real-world datasets have demonstrated the effectiveness of our model on the PRG task.

Top-N item recommendation has been a widely studied task from implicit feedback. Although much progress has been made with neural methods, there is increasing concern on appropriate evaluation of recommendation algorithms. In this paper, we revisit alternative experimental settings for evaluating top-N recommendation algorithms, considering three important factors, namely dataset splitting, sampled metrics and domain selection. We select eight representative recommendation algorithms (covering both traditional and neural methods) and construct extensive experiments on a very large dataset. By carefully revisiting different options, we make several important findings on the three factors, which directly provide useful suggestions on how to appropriately set up the experiments for top-N item recommendation.

Recently, significant progress has been made in sequential recommendation with deep learning. Existing neural sequential recommendation models usually rely on the item prediction loss to learn model parameters or data representations. However, the model trained with this loss is prone to suffer from data sparsity problem. Since it overemphasizes the final performance, the association or fusion between context data and sequence data has not been well captured and utilized for sequential recommendation.

Recently, conversational recommender system (CRS) has become an emerging and practical research topic. Most of the existing CRS methods focus on learning effective preference representations for users from conversation data alone. While, we take a new perspective to leverage historical interaction data for improving CRS. For this purpose, we propose a novel pre-training approach to integrating both item-based preference sequence (from historical interaction data) and attribute-based preference sequence (from conversation data) via pre-training methods. We carefully design two pre-training tasks to enhance information fusion between item- and attribute-based preference. To improve the learning performance, we further develop an effective negative sample generator which can produce high-quality negative samples. Experiment results on two real-world datasets have demonstrated the effectiveness of our approach for improving CRS.

Conversational recommender systems (CRS) aim to recommend high-quality items to users through interactive conversations.To develop an effective CRS, the support of high-quality datasets is essential.Existing CRS datasets mainly focus on immediate requests from users, while lack proactive guidance to the recommendation scenario.In this paper, we contribute a new CRS dataset named TG-ReDial (Recommendation through Topic-Guided Dialog).Our dataset has two major features.First, it incorporates topic threads to enforce natural semantic transitions towards the recommendation scenario.Second, it is created in a semi-automatic way, hence human annotation is more reasonable and controllable.Based on TG-ReDial, we present the task of topic-guided conversational recommendation, and propose an effective approach to this task.Extensive experiments have demonstrated the effectiveness of our approach on three sub-tasks, namely topic prediction, item recommendation and response generation.TG-ReDial is available at https

Road network is the core component of urban transportation, and it is widely useful in various traffic-related systems and applications. Due to its important role, it is essential to develop general, effective, and robust road network representation models. Although several efforts have been made in this direction, they cannot fully capture the complex characteristics of road networks.

Conversational recommender systems (CRS) aim to recommend high-quality items to users through interactive conversations. Although several efforts have been made for CRS, two major issues still remain to be solved. First, the conversation data itself lacks of sufficient contextual information for accurately understanding users' preference. Second, there is a semantic gap between natural language expression and item-level user preference.

Recently, deep learning has made significant progress in the task of sequential recommendation. Existing neural sequential recommenders typically adopt a generative way trained with Maximum Likelihood Estimation (MLE). When context information (called factor) is involved, it is difficult to analyze when and how each individual factor would affect the final recommendation performance.

For sequential recommendation, it is essential to capture and predict future or long-term user preference for generating accurate recommendation over time. To improve the predictive capacity, we adopt reinforcement learning (RL) for developing effective sequential recommenders. However, user-item interaction data is likely to be sparse, complicated and time-varying. It is not easy to directly apply RL techniques to improve the performance of sequential recommendation.

The task of Knowledge Graph Completion (KGC) aims to automatically infer the missing fact information in Knowledge Graph (KG). In this paper, we take a new perspective that aims to leverage rich user-item interaction data (user interaction data for short) for improving the KGC task. Our work is inspired by the observation that many KG entities correspond to online items in application systems. However, the two kinds of data sources have very different intrinsic characteristics, and it is likely to hurt the original performance using simple fusion strategy.

This article presents a network embedding approach to automatically generate tags for microblog users. Instead of using text data, we aim to annotate microblog users with meaningful tags by leveraging rich social link data. To utilize directed social links, we use two kinds of node representations for modeling user interest in terms of their followers and followees, respectively. To alleviate the sparsity problem, we propose a novel method based on two transformation functions for capturing implicit interest similarity. Different from previous works on capturing high-order proximity, our model is able to directly characterize the effect of the context user on the proximity of node pairs. Another novelty of our model is that the importance scores of users learned from the classic PageRank algorithm are utilized to set the link weights. By using such weights, our model is more capable of disentangling the interest similarity evidence of a link. We jointly consider the above factors when designing the final objective function. We construct a very large evaluation set consisting of 2.6M users, 0.5M tags, and 0.8B following links. To our knowledge, it is the largest reported dataset for microblog user tagging in the literature. Extensive experiments on this dataset demonstrate the effectiveness of the proposed approach. We implement this approach with several optimization techniques, which makes our model easy to scale to very large social networks. Ubiquitous social links provide important data resources to understand user interests. Our work provides an effective and efficient solution to annotate user interests solely using the link data, which has important practical value in industry. To illustrate the use of our models, we implement a demonstration system for visualizing, navigating, and searching microblog users.

Generating long and informative review text is a challenging natural language generation task.Previous work focuses on word-level generation, neglecting the importance of topical and syntactic characteristics from natural languages.In this paper, we propose a novel review generation model by characterizing an elaborately designed aspect-aware coarse-tofine generation process.First, we model the aspect transitions to capture the overall content flow.Then, to generate a sentence, an aspectaware sketch will be predicted using an aspectaware decoder.Finally, another decoder fills in the semantic slots by generating corresponding words.Our approach is able to jointly utilize aspect semantics, syntactic sketch, and context information.Extensive experiments results have demonstrated the effectiveness of the proposed model.

Fastest Route Recommendation (FRR) aims to find the fastest path in response to user's queries in a large complex road network. Early studies cast the FRR task as a pathfinding problem on graphs and adopt heuristic algorithms as the major solution due to the efficiency and robustness. A major problem of heuristic algorithms is that the heuristic function is usually empirically set with simple methods, which is difficult to model other useful factors. In this paper, we extend the classic A* algorithm for the FRR task by modeling complex traffic information with neural networks. Specially, we identify an important factor that is important to improve the FRR task, i.e. the estimation of travel time. For this purpose, we first develop a module for predicting the time-varying traffic speed for a road segment, which is the foundation for estimating the travel time. Conditioned on this module, we further design another module to estimate the fastest travel time between two locations connected by routes. We adopt neural networks to implement both modules for enabling the capacity of modeling complex traffic characteristics and dynamics. In this way, the original two cost functions of A* algorithm have been set in a more principled way with neural networks. To our knowledge, we are the first to use neural networks for improving A* algorithm in the FRR task. It elegantly combines the merits of A* algorithm and the powerful modeling capacities of neural networks for the FRR task. Extensive results on the three real-world datasets have shown the effectiveness and robustness of the proposed model.

Shuqing Bian, Wayne Xin Zhao, Yang Song, Tao Zhang, Ji-Rong Wen. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Siqing Li, Wayne Xin Zhao, Eddy Jing Yin, Ji-Rong Wen. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Personalized Route Recommendation (PRR) aims to generate user-specific route suggestions in response to users' route queries. Early studies cast the PRR task as a pathfinding problem on graphs, and adopt adapted search algorithms by integrating heuristic strategies. Although these methods are effective to some extent, they require setting the cost functions with heuristics. In addition, it is difficult to utilize useful context information in the search procedure. To address these issues, we propose using neural networks to automatically learn the cost functions of a classic heuristic algorithm, namely A* algorithm, for the PRR task. Our model consists of two components. First, we employ attention-based Recurrent Neural Networks (RNN) to model the cost from the source to the candidate location by incorporating useful context information. Instead of learning a single cost value, the RNN component is able to learn a time-varying vectorized representation for the moving state of a user. Second, we propose to use a value network for estimating the cost from a candidate location to the destination. For capturing structural characteristics, the value network is built on top of improved graph attention networks by incorporating the moving state of a user and other context information. The two components are integrated in a principled way for deriving a more accurate cost of a candidate location. Extensive experiment results on three real-world datasets have shown the effectiveness and robustness of the proposed model.

In e-commerce, users' demands are not only conditioned by their profile and preferences, but also by their recent purchases that may generate new demands, as well as periodical demands that depend on purchases made some time ago. We call them respectively short-term demands and long-term demands. In this paper, we propose a novel self-attentive Continuous-Time Recommendation model (CTRec) for capturing the evolving demands of users over time. For modeling such time-sensitive demands, a Demand-aware Hawkes Process (DHP) framework is designed in CTRec to learn from the discrete purchase records of users. More specifically, a convolutional neural network is utilized to capture the short-term demands; and a self-attention mechanism is employed to capture the periodical purchase cycles of long-term demands. All types of demands are fused in DHP to make final continuous-time recommendations. We conduct extensive experiments on four real-world commercial datasets to demonstrate that CTRec is effective for general sequential recommendation problems, including next-item and next-session/basket recommendations. We observe in particular that CTRec is capable of learning the purchase cycles of products and estimating the purchase time of a product given a user.

Recently, due to the unprecedented growth of multimedia data, cross-modal hashing has gained increasing attention for the efficient cross-media retrieval. Typically, existing methods on cross-modal hashing treat labels of one instance independently but overlook the correlations among labels. Indeed, in many real-world scenarios, like the online fashion domain, instances (items) are labeled with a set of categories correlated by certain hierarchy. In this paper, we propose a new end-to-end solution for supervised cross-modal hashing, named HiCHNet, which explicitly exploits the hierarchical labels of instances. In particular, by the pre-established label hierarchy, we comprehensively characterize each modality of the instance with a set of layer-wise hash representations. In essence, hash codes are encouraged to not only preserve the layer-wise semantic similarities encoded by the label hierarchy, but also retain the hierarchical discriminative capabilities. Due to the lack of benchmark datasets, apart from adapting the existing dataset FashionVC from fashion domain, we create a dataset from the online fashion platform Ssense consisting of 15,696 image-text pairs labeled by 32 hierarchical categories. Extensive experiments on two real-world datasets demonstrate the superiority of our model over the state-of-the-art methods.

In this paper, we focus on the task of sequential recommendation using taxonomy data. Existing sequential recommendation methods usually adopt a single vectorized representation for learning the overall sequential characteristics, and have a limited modeling capacity in capturing multi-grained sequential characteristics over context information. Besides, existing methods often directly take the feature vectors derived from context information as auxiliary input, which is difficult to fully exploit the structural patterns in context information for learning preference representations. To address above issues, we propose a novel Taxonomy-aware Multi-hop Reasoning Network, named TMRN, which integrates a basic GRU-based sequential recommender with an elaborately designed memory-based multi-hop reasoning architecture. For enhancing the reasoning capacity, we incorporate taxonomy data as structural knowledge to instruct the learning of our model. We associate the learning of user preference in sequential recommendation with the category hierarchy in the taxonomy. Given a user, for each recommendation, we learn a unique preference representation corresponding to each level in the taxonomy based on her/his overall sequential preference. In this way, the overall, coarse-grained preference representation can be gradually refined in different levels from general to specific, and we are able to capture the evolvement and refinement of user preference over the taxonomy, which makes our model highly explainable. Extensive experiments show that our proposed model is superior to state-of-the-art baselines in terms of both effectiveness and interpretability.

Presents corrections to author information from the paper, “Characterizing and predicting early reviewers for effective product marketing on e-commerce websites,” (Bai, T., et al), IEEE Trans. Knowl. Data Eng., vol. 30, no. 12, pp. 2271–2284, Dec. 2018.

Due to the flexibility in modelling data heterogeneity, heterogeneous information network (HIN) has been adopted to characterize complex and heterogeneous auxiliary data in recommender systems, calledHIN based recommendation. It is challenging to develop effective methods for HIN based recommendation in both extraction and exploitation of the information from HINs. Most of HIN based recommendation methods rely on path based similarity, which cannot fully mine latent structure features of users and items. In this paper, we propose a novel heterogeneous network embedding based approach for HIN based recommendation, called HERec. To embed HINs, we design a meta-path based random walk strategy to generate meaningful node sequences for network embedding. The learned node embeddings are first transformed by a set of fusion functions, and subsequently integrated into an extended matrix factorization (MF) model. The extended MF model together with fusion functions are jointly optimized for the rating prediction task. Extensive experiments on three real-world datasets demonstrate the effectiveness of the HERec model. Moreover, we show the capability of the HERec model for the cold-start problem, and reveal that the transformed embedding information from HINs can improve the recommendation performance.

Hypertext documents, such as web pages and academic papers, are of great importance in delivering information in our daily life.Although being effective on plain documents, conventional text embedding methods suffer from information loss if directly adapted to hyper-documents.In this paper, we propose a general embedding approach for hyper-documents, namely, hyperdoc2vec, along with four criteria characterizing necessary information that hyper-document embedding models should preserve.Systematic comparisons are conducted between hyperdoc2vec and several competitors on two tasks, i.e., paper classification and citation recommendation, in the academic paper domain.Analyses and experiments both validate the superiority of hyperdoc2vec to other models w.r.t. the four criteria.

Xiangyang Zhou, Lu Li, Daxiang Dong, Yi Liu, Ying Chen, Wayne Xin Zhao, Dianhai Yu, Hua Wu. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018.

Since heterogeneous information network (HIN) is able to integrate complex information and contain rich semantics, there is a surge of HIN based recommendation in recent years. Although existing methods have achieved performance improvement to some extent, they still face the following problems: how to extensively exploit and comprehensively explore the local and global information in HIN for recommendation. To address these issues, we propose a unified model LGRec to fuse local and global information for top-N recommendation in HIN. We firstly model most informative local neighbor information for users and items respectively with a co-attention mechanism. In addition, our model learns effective relation representations between users and items to capture rich information in HIN by optimizing a multi-label classification problem. Finally, we combine the two parts into an unified model for top-N recommendation. Extensive experiments on four real-world datasets demonstrate the effectiveness of the proposed model.

Many domain-specific websites host a profile page for each entity (e.g., locations on Foursquare, movies on IMDb, and products on Amazon), and users can post comments on it. When commenting on an entity, users often mention other entities for reference or comparison. Compared with web pages and tweets, disambiguating the mentioned entities in user comments has not received much attention. This paper investigates linking fine-grained locations in Foursquare comments. We demonstrate that the focal location, i.e., the location that a comment is posted on, provides rich contexts for linking. To exploit such information, we represent the Foursquare data in a graph, which includes locations, comments, and their relations. A probabilistic model named FocalLink is proposed to estimate the probability that a user mentions a location when commenting on a focal location, by following different kinds of relations. Experimental results show that FocalLink is consistently superior to different baselines.

Heterogeneous information network (HIN) has been widely adopted in recommender systems due to its excellence in modeling complex context information. Although existing HIN based recommendation methods have achieved performance improvement to some extent, they have two major shortcomings. First, these models seldom learn an explicit representation for path or meta-path in the recommendation task. Second, they do not consider the mutual effect between the meta-path and the involved user-item pair in an interaction. To address these issues, we develop a novel deep neural network with the co-attention mechanism for leveraging rich meta-path based context for top-N recommendation. We elaborately design a three-way neural interaction model by explicitly incorporating meta-path based context. To construct the meta-path based context, we propose to use a priority based sampling technique to select high-quality path instances. Our model is able to learn effective representations for users, items and meta-path based context for implementing a powerful interaction function. The co-attention mechanism improves the representations for meta-path based con- text, users and items in a mutual enhancement way. Extensive experiments on three real-world datasets have demonstrated the effectiveness of the proposed model. In particular, the proposed model performs well in the cold-start scenario and has potentially good interpretability for the recommendation results.

Next basket recommendation is a new type of recommendation, which recommends a set of items, or a basket, to the user. Purchase in basket is a common behavior of consumers. Recently, deep neural networks have been applied to model sequential transactions of baskets in next basket recommendation. However, current methods do not track the user's evolving appetite for items explicitly, and they ignore important item attributes such as product category. In this paper, we propose a novel Attribute-aware Neural Attentive Model (ANAM) to address these problems. ANAM adopts an attention mechanism to explicitly model user's evolving appetite for items, and utilizes a hierarchical architecture to incorporate the attribute information. In specific, ANAM utilizes a recurrent neural network to model the user's sequential behavior over time, and relays the user's appetite toward items and their attributes to next basket through attention weights shared across baskets on the two different hierarchies. Experiment results on two public datasets (ıe Ta-Feng and JingDong) demonstrate the effectiveness of our ANAM model for next basket recommendation.

With the revival of neural networks, many studies try to adapt powerful sequential neural models, ıe Recurrent Neural Networks (RNN), to sequential recommendation. RNN-based networks encode historical interaction records into a hidden state vector. Although the state vector is able to encode sequential dependency, it still has limited representation power in capturing complicated user preference. It is difficult to capture fine-grained user preference from the interaction sequence. Furthermore, the latent vector representation is usually hard to understand and explain. To address these issues, in this paper, we propose a novel knowledge enhanced sequential recommender. Our model integrates the RNN-based networks with Key-Value Memory Network (KV-MN). We further incorporate knowledge base (KB) information to enhance the semantic representation of KV-MN. RNN-based models are good at capturing sequential user preference, while knowledge-enhanced KV-MNs are good at capturing attribute-level user preference. By using a hybrid of RNNs and KV-MNs, it is expected to be endowed with both benefits from these two components. The sequential preference representation together with the attribute-level preference representation are combined as the final representation of user preference. With the incorporation of KB information, our model is also highly interpretable. To our knowledge, it is the first time that sequential recommender is integrated with external memories by leveraging large-scale KB information.

This article studies the problem of learning effective representations for Location-Based Social Networks (LBSN), which is useful in many tasks such as location recommendation and link prediction. Existing network embedding methods mainly focus on capturing topology patterns reflected in social connections, while check-in sequences, the most important data type in LBSNs, are not directly modeled by these models. In this article, we propose a representation learning method for LBSNs called as JRLM++, which models check-in sequences together with social connections. To capture sequential relatedness, JRLM++ characterizes two levels of sequential contexts, namely fine-grained and coarse-grained contexts. We present a learning algorithm tailored to the hierarchical architecture of the proposed model. We conduct extensive experiments on two important applications using real-world datasets. The experimental results demonstrate the superiority of our model. The proposed model can generate effective representations for both users and locations in the same embedding space, which can be further utilized to improve multiple LBSN tasks.

Online reviews have become an important source of information for users before making an informed purchase decision. Early reviews of a product tend to have a high impact on the subsequent product sales. In this paper, we take the initiative to study the behavior characteristics of early reviewers through their posted reviews on two real-world large e-commerce platforms, i.e., Amazon and Yelp. In specific, we divide product lifetime into three consecutive stages, namelyearly,majority, andlaggards. A user who has posted a review in the early stage is considered as an early reviewer. We quantitatively characterize early reviewers based on their rating behaviors, the helpfulness scores received from others and the correlation of their reviews with product popularity. We have found that (1) an early reviewer tends to assign a higher average rating score; and (2) an early reviewer tends to post more helpful reviews. Our analysis of product reviews also indicates that early reviewers’ ratings and their received helpfulness scores are likely to influence product popularity. By viewing the review posting process as a multiplayer competition game, we propose a novel margin-based embedding model for early reviewer prediction. Extensive experiments on two different e-commerce datasets have shown that our proposed approach outperforms a number of competitive baselines.

Many domain-specific websites host a profile page for each entity (e.g., locations on Foursquare, movies on IMDb, and products on Amazon) for users to post comments on. When commenting on an entity, users often mention other entities for reference or comparison. Compared with web pages and tweets, the problem of disambiguating the mentioned entities in user comments has not received much attention. This paper investigates linking fine-grained locations in Foursquare comments. We demonstrate that the focal location, i.e., the location that a comment is posted on, provides rich contexts for the linking task. To exploit such information, we represent the Foursquare data in a graph, which includes locations, comments, and their relations. A probabilistic model named FocalLink is proposed to estimate the probability that a user mentions a location when commenting on a focal location, by following different kinds of relations. Experimental results show that FocalLink is consistently superior under different collective linking settings.

Search result diversification aims to retrieve diverse results to satisfy as many different information needs as possible. Supervised methods have been proposed recently to learn ranking functions and they have been shown to produce superior results to unsupervised methods. However, these methods use implicit approaches based on the principle of Maximal Marginal Relevance (MMR). In this paper, we propose a learning framework for explicit result diversification where subtopics are explicitly modeled. Based on the information contained in the sequence of selected documents, we use the attention mechanism to capture the subtopics to be focused on while selecting the next document, which naturally fits our task of document selection for diversification. As a preliminary attempt, we employ recurrent neural networks and max pooling to instantiate the framework. We use both distributed representations and traditional relevance features to model documents in the implementation. The framework is flexible to model query intent in either a flat list or a hierarchy. Experimental results show that the proposed method significantly outperforms all the existing search result diversification approaches.

Online Judge (OJ) systems have been widely used in many areas, including programming, mathematical problems solving, and job interviews. Unlike other online learning systems, such as Massive Open Online Course, most OJ systems are designed for self-directed learning without the intervention of teachers. Also, in most OJ systems, problems are simply listed in volumes and there is no clear organization of them by topics or difficulty levels. As such, problems in the same volume are mixed in terms of topics or difficulty levels. By analyzing large-scale users’ learning traces, we observe that there are two major learning modes (or patterns). Users either practice problems in a sequential manner from the same volume regardless of their topics or they attempt problems about the same topic, which may spread across multiple volumes. Our observation is consistent with the findings in classic educational psychology. Based on our observation, we propose a novel two-mode Markov topic model to automatically detect the topics of online problems by jointly characterizing the two learning modes. For further predicting the difficulty level of online problems, we propose a competition-based expertise model using the learned topic information. Extensive experiments on three large OJ datasets have demonstrated the effectiveness of our approach in three different tasks, including skill topic extraction, expertise competition prediction and problem recommendation.

Recently, deep neural networks have been widely applied to recommender systems. A representative work is to utilize deep learning for modeling complex user-item interactions. However, similar to traditional latent factor models by factorizing user-item interactions, they tend to be ineffective to capture localized information. Localized information, such as neighborhood, is important to recommender systems in complementing the user-item interaction data. Based on this consideration, we propose a novel Neighborhood-based Neural Collaborative Filtering model (NNCF). To the best of our knowledge, it is the first time that the neighborhood information is integrated into the neural collaborative filtering methods. Extensive experiments on three real-world datasets demonstrate the effectiveness of our model for the implicit recommendation task.

Conventional correlated topic models are able to capture correlation structure among latent topics by replacing the Dirichlet prior with the logistic normal distribution. Word embeddings have been proven to be able to capture semantic regularities in language. Therefore, the semantic relatedness and correlations between words can be directly calculated in the word embedding space, for example, via cosine values. In this paper, we propose a novel correlated topic model using word embeddings. The proposed model enables us to exploit the additional word-level correlation information in word embeddings and directly model topic correlation in the continuous word embedding space. In the model, words in documents are replaced with meaningful word embeddings, topics are modeled as multivariate Gaussian distributions over the word embeddings and topic correlations are learned among the continuous Gaussian topics. A Gibbs sampling solution with data augmentation is given to perform inference. We evaluate our model on the 20 Newsgroups dataset and the Reuters-21578 dataset qualitatively and quantitatively. The experimental results show the effectiveness of our proposed model.

Search result diversification aims to retrieve diverse results to satisfy as many different information needs as possible. Supervised methods have been proposed recently to learn ranking functions and they have been shown to produce superior results to unsupervised methods. However, these methods use implicit approaches based on the principle of Maximal Marginal Relevance (MMR). In this paper, we propose a learning framework for explicit result diversification where subtopics are explicitly modeled. Based on the information contained in the sequence of selected documents, we use attention mechanism to capture the subtopics to be focused on while selecting the next document, which naturally fits our task of document selection for diversification. The framework is implemented using recurrent neural networks and max-pooling which combine distributed representations and traditional relevance features. Our experiments show that the proposed method significantly outperforms all the existing methods.

Cold-start recommendation is one of the most challenging problems in recommender systems. An important approach to cold-start recommendation is to conduct an interview for new users, called the interview-based approach . Among the interview-based methods, Representative-Based Matrix Factorization (RBMF) [24] provides an effective solution with appealing merits: it represents users over selected representative items, which makes the recommendations highly intuitive and interpretable. However, RBMF only utilizes a global set of representative items to model all users. Such a representation is somehow too strict and may not be flexible enough to capture varying users’ interests. To address this problem, we propose a novel interview-based model to dynamically create meaningful user groups using decision trees and then select local representative items for different groups. A two-round interview is performed for a new user. In the first round, l 1 global questions are issued for group division, while in the second round, l 2 local-group-specific questions are given to derive local representation. We collect the feedback on the (l 1 +l 2 ) items to learn the user representations. By putting these steps together, we develop a joint optimization model, named local representative-based matrix factorization , for new user recommendations. Extensive experiments on three public datasets have demonstrated the effectiveness of the proposed model compared with several competitive baselines.

Two characteristics of location-based services are mobile trajectories and the ability to facilitate social networking. The recording of trajectory data contributes valuable resources towards understanding users’ geographical movement behaviors. Social networking is possible when users are able to quickly connect to anyone nearby. A social network with location based services is known as location-based social network (LBSN). As shown in Cho et al. [2013], locations that are frequently visited by socially related persons tend to be correlated, which indicates the close association between social connections and trajectory behaviors of users in LBSNs. To better analyze and mine LBSN data, we need to have a comprehensive view of each of these two aspects, i.e., the mobile trajectory data and the social network. Specifically, we present a novel neural network model that can jointly model both social networks and mobile trajectories. Our model consists of two components: the construction of social networks and the generation of mobile trajectories. First we adopt a network embedding method for the construction of social networks: a networking representation can be derived for a user. The key to our model lies in generating mobile trajectories. Second, we consider four factors that influence the generation process of mobile trajectories: user visit preference, influence of friends, short-term sequential contexts, and long-term sequential contexts. To characterize the last two contexts, we employ the RNN and GRU models to capture the sequential relatedness in mobile trajectories at the short or long term levels. Finally, the two components are tied by sharing the user network representations. Experimental results on two important applications demonstrate the effectiveness of our model. In particular, the improvement over baselines is more significant when either network structure or trajectory data is sparse.

With the great popularity of mobile devices, the amount of mobile apps has grown at a more dramatic rate than ever expected. A technical challenge is how to recommend suitable apps to mobile users. In this work, we identify and focus on a unique characteristic that exists in mobile app recommendation—that is, an app usually corresponds to multiple release versions. Based on this characteristic, we propose a fine-grain version-aware app recommendation problem. Instead of directly learning the users’ preferences over the apps, we aim to infer the ratings of users on a specific version of an app. However, the user-version rating matrix will be sparser than the corresponding user-app rating matrix, making existing recommendation methods less effective. In view of this, our approach has made two major extensions. First, we leverage the review text that is associated with each rating record; more importantly, we consider two types of version-based correlations. The first type is to capture the temporal correlations between multiple versions within the same app, and the second type of correlation is to capture the aggregation correlations between similar apps. Experimental results on a large dataset demonstrate the superiority of our approach over several competitive methods.

The task of tweet timeline generation (TTG) aims at selecting a small set of representative tweets to generate a meaningful timeline and providing enough coverage for a given topical query. This paper presents an approach based on determinantal point processes (DPPs) by jointly modeling the topical relevance of each selected tweet and overall selectional diversity. Aiming at better treatment for balancing relevance and diversity, we introduce two novel strategies, namely spectral rescaling and topical prior. Extensive experiments on the public TREC 2014 dataset demonstrate that our proposed DPP model along with the two strategies can achieve fairly competitive results against the state-of-the-art TTG systems.

We present in this article an automated framework that extracts product adopter information from online reviews and incorporates the extracted information into feature-based matrix factorization for more effective product recommendation. In specific, we propose a bootstrapping approach for the extraction of product adopters from review text and categorize them into a number of different demographic categories. The aggregated demographic information of many product adopters can be used to characterize both products and users in the form of distributions over different demographic categories. We further propose a graph-based method to iteratively update user- and product-related distributions more reliably in a heterogeneous user--product graph and incorporate them as features into the matrix factorization approach for product recommendation. Our experimental results on a large dataset crawled from J ing D ong , the largest B2C e-commerce website in China, show that our proposed framework outperforms a number of competitive baselines for product recommendation.

In recent years, the boundaries between e-commerce and social networking have become increasingly blurred. Many e-commerce Web sites support the mechanism of social login where users can sign on the Web sites using their social network identities such as their Facebook or Twitter accounts. Users can also post their newly purchased products on microblogs with links to the e-commerce product Web pages. In this paper, we propose a novel solution for cross-site cold-start product recommendation, which aims to recommend products from e-commerce Web sites to users at social networking sites in “cold-start” situations, a problem which has rarely been explored before. A major challenge is how to leverage knowledge extracted from social networking sites for cross-site cold-start product recommendation. We propose to use the linked users across social networking sites and e-commerce Web sites (users who have social networking accounts and have made purchases on e-commerce Web sites) as a bridge to map users' social networking features to another feature representation for product recommendation. In specific, we propose learning both users' and products' feature representations (called user embeddings and product embeddings, respectively) from data collected from e-commerce Web sites using recurrent neural networks and then apply a modified gradient boosting trees method to transform users' social networking features into user embeddings. We then develop a feature-based matrix factorization approach which can leverage the learnt user embeddings for cold-start product recommendation. Experimental results on a large dataset constructed from the largest Chinese microblogging service Sina Weibo and the largest Chinese B2C e-commerce website JingDong have shown the effectiveness of our proposed framework.

The proliferation of location-based social networks, such as Foursquare and Facebook Places, offers a variety of ways to record human mobility, including user generated geo-tagged contents, check-in services, and mobile apps. Although trajectory data is of great value to many applications, it is challenging to analyze and mine trajectory data due to the complex characteristics reflected in human mobility, which is affected by multiple contextual information. In this paper, we propose a Multi-Context Trajectory Embedding Model, called MC-TEM, to explore contexts in a systematic way. MC-TEM is developed in the distributed representation learning framework, and it is flexible to characterize various kinds of useful contexts for different applications. To the best of our knowledge, it is the first time that the distributed representation learning methods apply to trajectory data. We formally incorporate multiple context information of trajectory data into the proposed model, including user-level, trajectory-level, location-level, and temporal contexts. All the context information is represented in the same embedding space. We apply MC-TEM to two challenging tasks, namely location recommendation and social link prediction. We conduct extensive experiments on three real-world datasets. Extensive experiment results have demonstrated the superiority of our MC-TEM model over several state-of-the-art methods.

With the pervasiveness of location-based social networks, it becomes increasingly important to consider the social characteristics of locations shared among persons. Several studies have been proposed to infer social strength by using trajectory similarity. However, these studies have two major shortcomings. First, they rely on the explicit co-occurrence of check-in locations. In this situation, a user pair of two friends who seldom share common locations or a user pair of two strangers who heavily share common visited locations will receive an unreliable estimation of the real social strength between them. Second, these studies do not consider how the overall trajectory patterns of users change with the varying of living styles. In this article, we propose a probabilistic generative model to mine latent lifestyle-related patterns from human trajectory data for inferring social strength. It can automatically learnfunctionality topicsconsisting of locations with similar service functions and transition probabilities over the set of functionality topics. Furthermore, a lifestyle is modeled as a unique transition probability matrix over the set of functionality topics. A user has a preference distribution over the set of lifestyles, and he or she is able to select over multiple lifestyles to adapt to different living contexts. The learned lifestyle-related patterns are subsequently used as features in a supervised learner for both strength estimation and link prediction. We conduct extensive experiments to evaluate the performance of the proposed method on two real-world datasets. The experimental results demonstrate the effectiveness of our proposed method.

In this paper, we propose to study the problem of identifying and classifying tweets into intent categories. For example, a tweet “I wanna buy a new car” indicates the user’s intent for buying a car. Identifying such intent tweets will have great commercial value among others. In particular, it is important that we can distinguish different types of intent tweets. We propose to classify intent tweets into six categories, namely Food & Drink, Travel, Career & Education, Goods & Services, Event and Activities and Trifle. We propose a semisupervised learning approach to categorizing intent tweets into the six categories.We construct a test collection by using a bootstrap method. Our experimental results show that our approach is effective in inferring intent categories for tweets.

In this paper, we explore the idea of social role theory (SRT) and propose a novel regularized topic model which incorporates SRT into the generative process of social media content. We assume that a user can play multiple social roles, and each social role serves to fulfil different duties and is associated with a role-driven distribution over latent topics. In particular, we focus on social roles corresponding to the most common social activities on social networks. Our model is instantiated on microblogs, i.e., Twitter and community question-answering (cQA), i.e., Yahoo!Answers, where social roles on Twitter include “originators” and “propagators”, and roles on cQA are “askers” and “answerers”. Both explicit and implicit interactions between users are taken into account and modeled as regularization factors. To evaluate the performance of our proposed method, we have conducted extensive experiments on two Twitter datasets and two cQA datasets. Furthermore, we also consider multi-role modeling for scientific papers where an author's research expertise area is considered as a social role. A novel application of detecting users' research interests through topical keyword labeling based on the results of our multi-role model has been presented. The evaluation results have shown the feasibility and effectiveness of our model.

Compression algorithms are important for data-oriented tasks, especially in the era of “Big Data.” Modern processors equipped with powerful SIMD instruction sets provide us with an opportunity for achieving better compression performance. Previous research has shown that SIMD-based optimizations can multiply decoding speeds. Following these pioneering studies, we propose a general approach to accelerate compression algorithms. By instantiating the approach, we have developed several novel integer compression algorithms, called Group-Simple, Group-Scheme, Group-AFOR, and Group-PFD, and implemented their corresponding vectorized versions. We evaluate the proposed algorithms on two public TREC datasets, a Wikipedia dataset, and a Twitter dataset. With competitive compression ratios and encoding speeds, our SIMD-based algorithms outperform state-of-the-art nonvectorized algorithms with respect to decoding speeds.

Existing approaches of social influence analysis usually focus on how to develop effective algorithms to quantize users' influence scores. They rarely consider a person's expertise levels which are arguably important to influence measures. In this paper, we propose a computational approach to measuring the correlation between expertise and social media influence, and we take a new perspective to understand social media influence by incorporating expertise into influence analysis. We carefully constructed a large dataset of 13,684 Chinese celebrities from Sina Weibo (literally “Sina microblogging”). We found that there is a strong correlation between expertise levels and social media influence scores. In addition, different expertise levels showed influence variation patterns: high-expertise celebrities have stronger influence on the “audience” in their expertise domains.

Product recommender systems are often deployed by e-commerce websites to improve user experience and increase sales. However, recommendation is limited by the product information hosted in those e-commerce sites and is only triggered when users are performing e-commerce activities. In this paper, we develop a novel product recommender system called METIS, a MErchanT Intelligence recommender System, which detects users' purchase intents from their microblogs in near real-time and makes product recommendation based on matching the users' demographic information extracted from their public profiles with product demographics learned from microblogs and online reviews. METIS distinguishes itself from traditional product recommender systems in the following aspects: 1) METIS was developed based on a microblogging service platform. As such, it is not limited by the information available in any specific e-commerce website. In addition, METIS is able to track users' purchase intents in near real-time and make recommendations accordingly. 2) In METIS, product recommendation is framed as a learning to rank problem. Users' characteristics extracted from their public profiles in microblogs and products' demographics learned from both online product reviews and microblogs are fed into learning to rank algorithms for product recommendation. We have evaluated our system in a large dataset crawled from Sina Weibo. The experimental results have verified the feasibility and effectiveness of our system. We have also made a demo version of our system publicly available and have implemented a live system which allows registered users to receive recommendations in real time.

Learning user interests from online social networks helps to better understand user behaviors and provides useful guidance to design user-centric applications. Apart from analyzing users' online content, it is also important to consider users' social connections in the social Web. Graph regularization methods have been widely used in various text mining tasks, which can leverage the graph structure information extracted from data. Previously, graph regularization methods operate under the cluster assumption that nearby nodes are more similar and nodes on the same structure (typically referred to as a cluster or a manifold) are likely to be similar. We argue that learning user interests from complex, sparse, and dynamic social networks should be based on the link structure assumption under which node similarities are evaluated based on the local link structures instead of explicit links between two nodes. We propose a regularization framework based on the relation bipartite graph, which can be constructed from any type of relations. Using Twitter as our case study, we evaluate our proposed framework from social networks built from retweet relations. Both quantitative and qualitative experiments show that our proposed method outperforms a few competitive baselines in learning user interests over a set of predefined topics. It also gives superior results compared to the baselines on retweet prediction and topical authority identification.

Compression algorithms have been quite important for data oriented tasks, especially in the era of Big Data. The rapid development of modern processors facilitates us with powerful SIMD instruction sets, which provides an opportunity for better performance. Although SIMD based optimization on compression have been explored in some studies [2, 7], these studies usually focus on modifying the existing algorithms to fit into the SIMD instruction. In this paper, we propose a compression framework with a novel storage layout format, which aims to improve instruction-level parallelizability of compression algorithms. By instantiating the framework, we design a novel compression algorithm family, called Group-Scheme, and present a parallelized version of Group-Scheme, called SIMD-Group-Scheme. We evaluate the proposed algorithms on two public TREC data sets. With very competitive performance on compression ratio and encoding speed, SIMD-Group-Scheme significantly outperforms the implementation without SIMD instructions and state-of-the-art algorithm (i.e. SIMD-G8IU [7]), w.r.t decoding speed.

A large number of studies have been devoted to modeling the contents and interactions between users on Twitter. In this paper, we propose a method inspired from Social Role Theory (SRT), which assumes that a user behaves differently with different roles in the generation process of Twitter content. We consider the two most distinctive social roles on Twitter: originator and propagator, who respectively posts original messages and retweets or forwards the messages from others. In addition, we also consider role-specific social interactions, especially implicit interactions between users who share some common interests. All the above elements are integrated into a novel regularized topic model. We evaluate the proposed method on real Twitter data. The results show that our method is more effective than the existing ones which do not distinguish social roles.

Hot trends are likely to bring new business opportunities.For example, "Air Pollution" might lead to a significant increase of the sales of related products, e.g., mouth mask.For ecommerce companies, it is very important to make rapid and correct response to these hot trends in order to improve product sales.In this paper, we take the initiative to study the task of how to identify trend related products.The major novelty of our work is that we automatically learn commercial intents revealed from microblogs.We carefully construct a data collection for this task and present quite a few insightful findings.In order to solve this problem, we further propose a graph based method, which jointly models relevance and associativity.We perform extensive experiments and the results showed that our methods are very effective.

Timeline generation is an important research task which can help users to have a quick understanding of the overall evolution of any given topic. It thus attracts much attention from research communities in recent years. Nevertheless, existing work on timeline generation often ignores an important factor, the attention attracted to topics of interest (hereafter termed "social attention"). Without taking into consideration social attention, the generated timelines may not reflect users' collective interests. In this paper, we study how to incorporate social attention in the generation of timeline summaries. In particular, for a given topic, we capture social attention by learning users' collective interests in the form of word distributions from Twitter, which are subsequently incorporated into a unified framework for timeline summary generation. We construct four evaluation sets over six diverse topics. We demonstrate that our proposed approach is able to generate both informative and interesting timelines. Our work sheds light on the feasibility of incorporating social attention into traditional text mining tasks.

We present a novel graph-based framework for timeline summarization, the task of creating different summaries for different timestamps but for the same topic. Our work extends timeline summarization to a multimodal setting and creates timelines that are both textual and visual. Our approach exploits the fact that news documents are often accompanied by pictures and the two share some common content. Our model optimizes local summary creation and global timeline generation jointly following an iterative approach based on mutual reinforcement and co-ranking. In our algorithm, individual summaries are generated by taking into account the mutual dependencies between sentences and images, and are iteratively refined by considering how they contribute to the global timeline and its coherence. Experiments on real-world datasets show that the timelines produced by our model outperform several competitive baselines both in terms of ROUGE and when assessed by human evaluators.

We present EventSearch, a system for event extraction and retrieval on four types of news-related historical data, i.e., Web news articles, newspapers, TV news program, and micro-blog short messages. The system incorporates over 11 million web pages extracted from "Web InfoMall", the Chinese Web Archive since 2001. The newspaper and TV news video clips also span from 2001 to 2011. The system, upon a user query, returns a list of event snippets from multiple data sources. A novel burst model is used to discover events from time-stamped texts. In addition to offline event extraction, our system also provides online event extraction to further meet the user needs. EventSearch provides meaningful analytics that synthesize an accurate description of events. Users interact with the system by ranking the identified events using different criteria (scale, recency and relevance) and submitting their own information needs in different input fields.

A large proportion of search engine queries contain phrases,namely a sequence of adjacent words. In this paper, we propose to use flat position index (a.k.a schema-independent index) for phrase query evaluation. In the flat position index, the entire document collection is viewed as a huge sequence of tokens. Each token is represented by one flat position, which is a unique position offset from the beginning of the collection. Each indexed term is associated with a list of the flat positions about that term in the sequence. To recover DocID from flat positions efficiently, we propose a novel cache sensitive look-up table (CSLT), which is much faster than existing search algorithms. Experiments on TREC GOV2 data collection show that flat position index can reduce the index size and speed up phrase querying substantially, compared with traditional word-level index.

Various click models have been recently proposed as a principled approach to infer the relevance of documents from the clickthrough data. The inferred document relevance is potentially useful in evaluating the Web retrieval systems. In practice, it generally requires to acquire the accurate evaluation results within minimal users' query submissions. This problem is important for speeding up search engine development and evaluation cycle and acquiring reliable evaluation results on tail queries. In this paper, we propose a reordering framework for efficient evaluation problem in the context of clickthrough based Web retrieval evaluation. The main idea is to move up the documents that contribute more for the evaluation task. In this framework, we propose four intuitions and formulate them as an optimization problem. Both user study and TREC data based experiments validate that the reordering framework results in much fewer query submissions to get accurate evaluation results with only a little harm to the users' utility.

Bursty features in text streams are very useful in many text mining applications. Most existing studies detect bursty features based purely on term frequency changes without taking into account the semantic contexts of terms, and as a result the detected bursty features may not always be interesting or easy to interpret. In this paper we propose to model the contexts of bursty features using a language modeling approach. We then propose a novel topic diversity-based metric using the context models to find newsworthy bursty features. We also propose to use the context models to automatically assign meaningful tags to bursty features. Using a large corpus of a stream of news articles, we quantitatively show that the proposed context language models for bursty features can effectively help rank bursty features based on their newsworthiness and to assign meaningful tags to annotate bursty features.