Generative recommendation (GR) is an emerging paradigm that tokenizes items into discrete tokens and learns to autoregressively generate the next tokens as predictions. While this token-generation paradigm is expected to surpass traditional transductive methods, potentially generating new items directly based on semantics, we empirically show that GR models predominantly generate items seen during training and struggle to recommend unseen items. In this paper, we propose SpecGR, a plug-and-play framework that enables GR models to recommend new items in an inductive setting. SpecGR uses a drafter model with inductive capability to propose candidate items, which may include both existing items and new items. The GR model then acts as a verifier, accepting or rejecting candidates while retaining its strong ranking capabilities. We further introduce the guided re-drafting technique to make the proposed candidates more aligned with the outputs of generative recommendation models, improving the verification efficiency. We consider two variants for drafting: (1) using an auxiliary drafter model for better flexibility, or (2) leveraging the GR model's own encoder for parameter-efficient self-drafting. Extensive experiments on three real-world datasets demonstrate that SpecGR exhibits both strong inductive recommendation ability and the best overall performance among the compared methods.

Yupeng Hou, Jiacheng Li, Xiangjun Fu, Zhankui He, An Yan, Xiusi Chen, Julian McAuley. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Gagan Mundada, Rohan Surana, Nandhini Swaminathan, Bodhisattwa Prasad Majumder, Junda Wu, Julian McAuley, Zhouhang Xie. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Chuhan Wang, Xintong Li, Jennifer Yuntong Zhang, Junda Wu, Chengkai Huang, Lina Yao, Julian McAuley, Jingbo Shang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Rui Wang, Junda Wu, Yu Xia, Tong Yu, Ruiyi Zhang, Ryan A. Rossi, Subrata Mitra, Lina Yao, Julian McAuley. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

In this paper, we present ZoRRO (Zero-Weight Personalized Recommender System), a zero-weight and training-free framework for personalized news recommendation designed for scalable real-world deployment. We show that ZoRRO outperforms strong neural baselines in offline ranking evaluations and delivers click-through rate performance in online A/B testing that is nearly on par with a state-of-the-art deep learning model, while operating more than (600x) faster. Our experiments reveal gaps between offline and online performance, and show that models with similar click-through rate (CTR) outcomes can produce markedly different recommendation distributions, influencing the overall news flow. These findings position ZoRRO as a practical and efficient solution for large-scale news recommendation and highlight the importance of evaluating recommender systems using metrics beyond accuracy alone. Our code is available at https://github.com/johanneskruse/zorro.

Natural language information needs over symbolic music scores rarely reduce to a single-step lookup. Many queries require compositional Music Information Retrieval (MIR) that extracts multiple pieces of evidence from structured notation and aggregates them to answer the question. This setting remains challenging for Large Language Models due to the mismatch between natural language intents and symbolic representations, as well as the difficulty of reliably handling long structured contexts. Existing benchmarks only partially capture these retrieval demands, often emphasizing isolated theoretical knowledge or simplified settings. We introduce CSyMR-Bench, a benchmark for compositional MIR in symbolic music reasoning grounded in authentic user scenarios. It contains 126 multiple-choice questions curated from community discussions and professional examinations, where each item requires chaining multiple atomic analyses over a score to derive implicit musical evidence. To support diagnosis, we provide a taxonomy with six query intent categories and six analytical dimension tags. We further propose a tool-augmented retrieval and reasoning framework, CSyMR-Agent, that integrates a ReAct-style controller with deterministic symbolic analysis operators built with music21. Experiments across prompting baselines and agent variants show that tool-grounded compositional retrieval consistently outperforms Large Language Model-only approaches, yielding 5-7% absolute accuracy gains, with the largest improvements on analysis-heavy categories.

Recent advancements in diffusion models have shown promising results in sequential recommendation (SR). Existing approaches predominantly rely on implicit conditional diffusion models, which compress user behaviors into a single representation during the forward diffusion process. While effective to some extent, this oversimplification often leads to the loss of sequential and contextual information, which is critical for understanding user behavior. Moreover, explicit information, such as user-item interactions or sequential patterns, remains underutilized, despite its potential to directly guide the recommendation process and improve precision. However, combining implicit and explicit information is non-trivial, as it requires dynamically integrating these complementary signals while avoiding noise and irrelevant patterns within user behaviors. To address these challenges, we propose Dual Conditional Diffusion Models for Sequential Recommendation (DCRec), which effectively integrates implicit and explicit information by embedding dual conditions into both the forward and reverse diffusion processes. This allows the model to retain valuable sequential and contextual information while leveraging explicit user-item interactions to guide the recommendation process. Specifically, we introduce the Dual Conditional Diffusion Transformer (DCDT), which dynamically integrate both implicit and explicit signals throughout the diffusion stages, ensuring contextual understanding and minimizing the influence of irrelevant patterns. Extensive experiments on public benchmark datasets demonstrate that DCRec significantly outperforms state-of-the-art methods.

Building personalized recommender systems and search experiences is a cornerstone of the modern data mining and applied machine learning (ML) community. Modern online platforms have a confluence of data including user-item interaction graphs, user and item-associated semantics (text, visual content, etc.), and metadata. Recent advancements in generative models and semantic encoders via large language models (LLMs), visual and audio encoders have significantly impacted research in relevant domains, enabling new directions in knowledge discovery and ability of models to better incorporate semantic context. These techniques are quickly advancing in the academic sphere, and adoption in industrial environments is growing. These advances force large questions about the future of search, recommendation and personalized experiences in the future. This workshop bridges the research gap between the use of generative models and recommendation for personalized systems. We will focus on topics spanning the interplay between such models and conventional personalized systems. Building upon the momentum of previous successful forums, we seek to engage a diverse audience from academia and industry, fostering a dialogue that incorporates fresh insights and anticipates over 100 attendees, including key stakeholders in the field.

Agentic recommendations cast recommenders as large language model (LLM) agents that can plan, reason, use tools, and interact with users of varying preferences in web applications. However, most existing agentic recommender systems focus on generic single-agent plan-execute workflows or multi-agent task decomposition pipelines. Without recommendation-oriented design, they often underuse the collaborative signals in the user–item interaction history, leading to unsatisfying recommendation results. To address this, we propose the Multi-Agent Collaborative Filtering (MACF) framework for agentic recommendations, drawing an analogy between traditional collaborative filtering algorithms and LLM-based multi-agent collaboration. Specifically, given a target user and query, we instantiate similar users and relevant items as LLM agents with unique profiles. Each agent is able to call retrieval tools, suggest candidate items, and interact with other agents. Different from the static preference aggregation in traditional collaborative filtering, MACF employs a central orchestrator agent to adaptively manage the collaboration between user and item agents via dynamic agent recruitment and personalized collaboration instruction. Experimental results on datasets from three different domains show the advantages of our MACF framework compared to strong agentic recommendation baselines.

Content recommender systems (RSs), which recommend content (e.g., news, videos) to users, can inadvertently facilitate the spread of misinformation (e.g., fake news, inaccurate content) due to their ignorance of content credibility. The widespread misinformation can lead to serious consequences, including public confusion and social unrest. Despite the pressing need to mitigate misinformation in content RSs, only a few studies have attempted to tackle this. Moreover, they generally require retraining the entire model from scratch, which is time-consuming and impractical for real-world applications. To this end, we propose MisEraser , a novel misinformation unlearning framework that effectively mitigates misinformation in content RSs without retraining. Specifically, MisEraser consists of three key components: (1) A misinformation disentanglement network , which effectively separates misinformation-specific information from item content while preserving misinformation-free information to maintain recommendation accuracy; (2) A misinformation fusion network , which fuses misinformation-specific signals from both users’ interaction sequences and the entire misinformation item set to enhance the comprehensiveness of misinformation unlearning; and (3) A misinformation erasing network , which effectively removes the fused misinformation-specific signals from trained recommendation models, enabling them to generate misinformation-suppressed and more responsible recommendations. Extensive experiments demonstrate the effectiveness of MisEraser in mitigating misinformation while maintaining high recommendation accuracy and efficiency.

Yu Xia, Subhojyoti Mukherjee, Zhouhang Xie, Junda Wu, Xintong Li, Ryan Aponte, Hanjia Lyu, Joe Barrow, Hongjie Chen, Franck Dernoncourt, Branislav Kveton, Tong Yu, Ruiyi Zhang, Jiuxiang Gu, Nesreen K. Ahmed, Yu Wang, Xiang Chen, Hanieh Deilamsalehy, Sungchul Kim, Zhengmian Hu, Yue Zhao, Nedim Lipka, Seunghyun Yoon, Ting-Hao Kenneth Huang, Zichao Wang, Puneet Mathur, Soumyabrata Pal, Koyel Mukherjee, Zhehao Zhang, Namyong Park, Thien Huu Nguyen, Jiebo Luo, Ryan A. Rossi, Julian McAuley. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

The emergence of large generative models is transforming the landscape of recommender systems. One of the most fundamental components in building these models is action tokenization, the process of converting human-readable data (e.g., user-item interactions) into machine-readable formats (e.g., discrete token sequences). In this tutorial, we present a comprehensive overview of existing action tokenization techniques, converting actions to (1) item IDs, (2) textual descriptions, and (3) semantic IDs. We then make an in-depth discussion on the challenges and open questions of building large generative recommendation models from the perspective of action tokenization. Materials of this tutorial are available at: https://large-genrec.github.io/.

Retrieval-augmented large language models (LLMs) have been remarkably competent in various NLP tasks. However, it was observed by previous works that retrieval is not always helpful, especially when the LLM is already knowledgable on the query to answer. Motivated by this, Adaptive Retrieval-Augmented Generation (ARAG) studies retrieving only when the knowledge asked by the query is absent in the LLM. Previous works of ARAG either require accessing the pre-training corpus or prompting with additional model inferences. Aiming to avoid such drawbacks, we propose to determine whether the model is knowledgeable on a query via inspecting the (contextualized) pre-trained token embeddings of LLMs. We hypothesize that such embeddings capture rich information on the model’s intrinsic knowledge base, which enables an efficient way of judging the necessity to retrieve from an external corpus. Extensive experiments demonstrate our ARAG approach’s superior performance across various benchmarks.

Chain-of-Thought (CoT) has been a widely adopted prompting method, eliciting impressive reasoning abilities of Large Language Models (LLMs). Inspired by the sequential thought structure of CoT, a number of Chain-of-X (CoX) methods have been developed to address challenges across diverse domains and tasks. In this paper, we provide a comprehensive survey of Chain-of-X methods for LLMs in different contexts. Specifically, we categorize them by taxonomies of nodes, i.e., the X in CoX, and application tasks. We also discuss the findings and implications of existing CoX methods, as well as potential future directions. Our survey aims to serve as a detailed and up-to-date resource for researchers seeking to apply the idea of CoT to broader scenarios.

We propose a symbolic generative task description language and a corresponding inference engine capable of representing arbitrary multimodal tasks as structured symbolic flows. Unlike conventional generative models that rely on large-scale training and implicit neural representations to learn cross-modal mappings—often at high computational cost and with limited flexibility—our framework introduces an explicit symbolic representation comprising three core primitives: $\color{blue}{\text{functions}}$, $\color{green}{\text{parameters}}$, and $\color {Purple}{\text{topological}}\,{\text{logic}}$. Leveraging a pre-trained language model, our inference engine maps natural language instructions directly to symbolic workflows in a training-free manner. Our framework successfully performs over 12 diverse multimodal generative tasks, demonstrating strong performance and flexibility without the need for task-specific tuning. Experiments show that our method not only matches or outperforms existing state-of-the-art unified models in content quality, but also offers greater efficiency, editability, and interruptibility. We believe that symbolic task representations provide a cost-effective and extensible foundation for advancing the capabilities of generative AI.

Image Difference Captioning (IDC) aims to generate natural language descriptions that highlight subtle differences between two visually similar images.While recent advances leverage pre-trained vision-language models to align fine-grained visual differences with textual semantics, existing supervised approaches often overly focus on dataset-specific language patterns and fail to capture fine-grained and context-aware preferences on IDC, due to limited annotation diversity and a lack of semantically informative negative examples during training, To address these limitations, we propose an adversarial direct preference optimization (ADPO) framework for IDC, which formulates IDC as a preference optimization problem under the Bradley-Terry-Luce model, directly aligning the captioning policy with pairwise difference preferences via Direct Preference Optimization (DPO).To model more accurate and diverse IDC preferences, we introduce an adversarially trained hard negative retriever that selects counterfactual captions, This results in a minimax optimization problem, which we solve via policy-gradient reinforcement learning, enabling the policy and retriever to improve jointly.By dynamically generating semantically challenging negatives, our method reduces reliance on dataset-specific patterns.Experiments on benchmark IDC datasets show that our approach outperforms existing baselines, especially in generating fine-grained and accurate difference descriptions.

Xintong Li, Junda Wu, Tong Yu, Rui Wang, Yu Wang, Xiang Chen, Jiuxiang Gu, Lina Yao, Julian McAuley, Jingbo Shang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across various visionlanguage tasks.However, their reasoning abilities in the multimodal symbolic music domain remain largely unexplored.We introduce Wild-Score, the first in-the-wild multimodal symbolic music reasoning and analysis benchmark, designed to evaluate MLLMs' capacity to interpret real-world music scores and answer complex musicological queries.Each instance in WildScore is sourced from genuine musical compositions and accompanied by authentic user-generated questions and discussions, capturing the intricacies of practical music analysis.To facilitate a comprehensive evaluation, we propose a systematic taxonomy, comprising both high-level and fine-grained musicological ontologies.Furthermore, we frame complex music reasoning as multiple-choice question answering, enabling controlled and scalable assessment of MLLMs' symbolic music understanding.Empirical benchmarking of state-ofthe-art MLLMs on WildScore reveals intriguing patterns in their visual-symbolic reasoning, uncovering both promising directions and persistent challenges for MLLMs in symbolic music reasoning and analysis.We release the dataset 1 and code 2 .

Large Language Model (LLM) agents are commonly tuned with supervised finetuning on ReAct-style expert trajectories or preference optimization over pairwise rollouts.Most of these methods focus on imitating specific expert behaviors or promoting chosen reasoning thoughts and actions over rejected ones.However, without reasoning and comparing over alternative actions, LLM agents finetuned with these methods may over-commit towards seemingly plausible but suboptimal actions due to limited action space exploration.To address this, in this paper we propose Self-taught Ac-tioN Deliberation (SAND) framework, enabling LLM agents to explicitly deliberate over candidate actions before committing to one.To tackle the challenges of when and what to deliberate given large action space and step-level action evaluation, we incorporate self-consistency action sampling and execution-guided action critique to help synthesize step-wise action deliberation thoughts using the base model of the LLM agent.In an iterative manner, the deliberation trajectories are then used to finetune the LLM agent itself.Evaluating on two representative interactive agent tasks, SAND achieves an average 20% improvement over supervised finetuning on initial expert data and also outperforms state-of-the-art agent tuning approaches.

Dayu Yang, Tianyang Liu, Daoan Zhang, Antoine Simoulin, Xiaoyi Liu, Yuwei Cao, Zhaopu Teng, Xin Qian, Grey Yang, Jiebo Luo, Julian McAuley. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

The recent explosion of generative AI-Music systems has raised numerous concerns over data copyright, licensing music from musicians, and the conflict between open-source AI and large prestige companies. Such issues highlight the need for publicly available, copyright-free musical data, in which there is a large shortage, particularly for symbolic music data. To alleviate this issue, we present PDMX: a large-scale open-source dataset of over 250K public domain MusicXML scores collected from the score-sharing forum MuseScore, making it the largest available copyright-free symbolic music dataset to our knowledge. PDMX additionally includes a wealth of both tag and user interaction metadata, allowing us to efficiently analyze the dataset and filter for high quality user-generated scores. Given the additional metadata afforded by our data collection process, we conduct multitrack music generation experiments evaluating how different representative subsets of PDMX lead to different behaviors in downstream models, and how user-rating statistics can be used as an effective measure of data quality.

Modeling temporal characteristics plays a significant role in the representation learning of audio waveform. We propose Contrastive Long-form Language-Audio Pretraining (CoLLAP) to significantly extend the perception window for both the input audio (up to 5 minutes) and the language descriptions (exceeding 250 words), while enabling contrastive learning across modalities and temporal dynamics. Leveraging recent Music-LLMs to generate long-form music captions for full-length songs, augmented with musical temporal structures, we collect 51.3K audio-text pairs derived from the large-scale AudioSet training dataset, where the average audio length reaches 288 seconds. We propose a novel contrastive learning architecture that fuses language representations with structured audio representations by segmenting each song into clips and extracting their embeddings. With an attention mechanism, we capture multimodal temporal correlations, allowing the model to automatically weigh and enhance the final fusion score for improved contrastive alignment. Finally, we develop two variants of the CoLLAP model with different types of backbone language models. Through comprehensive experiments on multiple long-form music-text retrieval datasets, we demonstrate consistent performance improvement in retrieval accuracy compared with baselines. We also show the pretrained CoLLAP models can be transferred to various music information retrieval tasks, with heterogeneous long-form multimodal contexts.

Recent music large language models (music LLMs) have shown great potential in music understanding through large-scale multimodal pre-training. While some existing music LLMs have been augmented with temporally-aware music captions, music information retrieval (MIR) features conventionally do not exist in music caption datasets, thus neglected by music-LLMs. To bridge the gap between recent music LLMs and conventional music information retrieval tasks, we propose FUTGA-MIR (Fine-grained Music Understanding through Temporally-enhanced Generative Augmentation with Music Information Retrieval) to enhance the existing music LLMs by augmenting them with MIR features and aligning with human feedback. We calibrate the original positional distribution of music clips in the created pre-training synthetic music captions, conditioned on MIR features. The redistributed synthetic dataset better respects realistic MIR distributions of music and is tightly aligned with realistic music captions as well. In addition, We incorporate human-annotated music captions and MIR features from the Harmonixset dataset for an additional fine-tuning step, which addresses sim-to-real domain gaps and improves performance on realistic music tracks. We further evaluate FUTGA-MIR on several downstream tasks, including music classification, retrieval, and generation, and demonstrate the versatile capacities of FUTGA-MIR and better generation quality compared with previous music captioning models. Generated temporal-aware music descriptions are illustrated in our demonstration https://namburiamit.github.io/futga-music.github.io/.

Despite significant advancements in large language models (LLMs), the rapid and frequent integration of small-scale experiences, such as interactions with sur- rounding objects, remains a substantial challenge. Two critical factors in assimilating these experiences are (1) **Efficacy**: the ability to accurately remember recent events; (2) **Retention**: the capacity to recall long-past experiences. Current methods either embed experiences within model parameters using continual learning, model editing, or knowledge distillation techniques, which often struggle with rapid updates and complex interactions, or rely on external storage to achieve long-term retention, thereby increasing storage requirements. In this paper, we propose **SELF-PARAM** (Self-Updatable Large Language Models with Parameter Integration). SELF-PARAM requires no extra parameters while ensuring near-optimal efficacy and long-term retention. Our method employs a training objective that minimizes the Kullback-Leibler (KL) divergence between the predictions of an original model (with access to contextual information) and a target model (without such access). By generating diverse question-answer pairs related to the knowledge and minimizing the KL divergence across this dataset, we update the target model to internalize the knowledge seamlessly within its parameters. Evaluations on question-answering and conversational recommendation tasks demonstrate that SELF-PARAM significantly outperforms existing methods, even when accounting for non-zero storage requirements. This advancement paves the way for more efficient and scalable integration of experiences in large language models by embedding knowledge directly into model parameters.

Large language models show impressive abilities in memorizing world knowledge, which leads to concerns regarding memorization of private information, toxic or sensitive knowledge, and copyrighted content. We introduce the problem of Large Scale Knowledge Washing, focusing on unlearning an extensive amount of factual knowledge. Previous unlearning methods usually define the reverse loss and update the model via backpropagation, which may affect the model's fluency and reasoning ability or even destroy the model due to extensive training with the reverse loss. Existing works introduce additional data from downstream tasks to prevent the model from losing capabilities, which requires downstream task awareness. Controlling the tradeoff of unlearning existing knowledge while maintaining existing capabilities is also challenging. To this end, we propose LaW (Large Scale Washing), where we update the MLP layers in decoder-only large language models to perform knowledge washing, as inspired by model editing methods. We derive a new objective with the knowledge to be unlearned to update the weights of certain MLP layers. Experimental results demonstrate the effectiveness of LaW in forgetting target knowledge while maximally maintaining reasoning ability. The code will be open-sourced.

Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than the comparable SOTA model) — the fastest TTM to our knowledge.

Offline evaluation of LLMs is crucial in understanding their capacities, though current methods remain underexplored in existing research. In this work, we focus on the offline evaluation of the chain-of-thought capabilities and show how to optimize LLMs based on the proposed evaluation method. To enable offline feedback with rich knowledge and reasoning paths, we use knowledge graphs (KGs) (e.g., Wikidata5M) to provide feedback on the generated chain of thoughts. Due to the heterogeneity between LLM reasoning and KG structures, direct interaction and feedback from knowledge graphs on LLM behavior are challenging, as they require accurate entity linking and grounding of LLM-generated chains of thought in the KG. To address the above challenge, we propose an offline chain-of-thought evaluation framework, OCEAN, which models chain-of-thought reasoning in LLMs as a Markov Decision Process (MDP), and evaluate the policy’s alignment with KG preference modeling. To overcome the reasoning heterogeneity and grounding problems, we leverage on-policy KG exploration and reinforcement learning to model a KG policy that generates token-level likelihood distributions for LLM-generated chain-of-thought reasoning paths, simulating KG reasoning preference. Then we incorporate the knowledge-graph feedback on the validity and alignment of the generated reasoning paths into inverse propensity scores and propose KG-IPS estimator. Theoretically, we prove the unbiasedness of the proposed KG-IPS estimator and provide a lower bound on its variance. With the off-policy evaluated value function, we can directly enable off-policy optimization to further enhance chain-of-thought alignment. Our empirical study shows that OCEAN can be efficiently optimized for generating chain-of-thought reasoning paths with higher estimated values without affecting LLMs’ general abilities in downstream tasks or their internal knowledge.

Teasers are an effective tool for promoting content in entertainment, commercial and educational fields. However, creating an effective teaser for long videos is challenging for it requires long-range multimodal modeling capability for the input videos, while necessitating maintaining audiovisual alignments, managing scene transitions and preserving factual accuracy for the output teasers. Due to the lack of a publicly-available dataset, progress along this research direction has been hindered. In this work, we present DocumentaryNet, a collection of 1,269 documentaries paired with their teasers, featuring multimodal data streams of video, speech, music, sound effects and narrations. With DocumentaryNet, we propose a new two-stage system for generating teasers from long documentaries. The proposed TeaserGen system first generates the teaser narration from the transcribed narration from the documentary using a pretrained large language model, and then selects the most relevant visual content to accompany the generated narration through language-vision models. For narration-video matching, we explore two approaches: a pretraining-based model using pretrained contrastive language-vision models and a deep sequential model that learns the mapping between the narrations and visuals. Our experimental results show that the pretraining-based approach is more effective at identifying relevant visual content than directly trained deep autoregressive models.

Equipping large language models (LLMs) with latent-space memory has attracted increasing attention as they can extend the context window of existing language models. However, retaining information from the distant past remains a challenge. For example, MemoryLLM (Wang et al., 2024a), as a representative work with latent-space memory, compresses past information into hidden states across all layers, forming a memory pool of 1B parameters. While effective for sequence lengths up to 16k tokens, it struggles to retain knowledge beyond 20k tokens. In this work, we address this limitation by introducing M+, a memory-augmented model based on MemoryLLM that significantly enhances long-term information retention. M+ integrates a long-term memory mechanism with a co-trained retriever, dynamically retrieving relevant information during text generation. We evaluate M+ on diverse benchmarks, including long-context understanding and knowledge retention tasks. Experimental results show that M+ significantly outperforms MemoryLLM and recent strong baselines, extending knowledge retention from under 20k to over 160k tokens with similar GPU memory overhead.

Generative recommendation (GR) is an emerging paradigm where user actions are tokenized into discrete token patterns and autoregressively generated as predictions. However, existing GR models tokenize each action independently, assigning the same fixed tokens to identical actions across all sequences without considering contextual relationships. This lack of context-awareness can lead to suboptimal performance, as the same action may hold different meanings depending on its surrounding context. To address this issue, we propose ActionPiece to explicitly incorporate context when tokenizing action sequences. In ActionPiece, each action is represented as a set of item features. Given the action sequence corpora, we construct the vocabulary by merging feature patterns as new tokens, based on their co-occurrence frequency both within individual sets and across adjacent sets. Considering the unordered nature of feature sets, we further introduce set permutation regularization, which produces multiple segmentations of action sequences with the same semantics. Our code is available at: https://github.com/google-deepmind/action_piece.

Visual Language Navigation (VLN) is a fundamental task within the field of Embodied AI, focusing on the ability of agents to navigate complex environments based on natural language instructions. Despite the progress made by existing methods, these methods often present some common challenges. First, they rely on pre-trained backbone models for visual perception, which struggle with the dynamic viewpoints in VLN scenarios. Second, the performance is limited when using pre-trained LLMs or VLMs without fine-tuning, due to the absence of VLN domain knowledge. Third, while fine-tuning LLMs and VLMs can improve results, their computational costs are higher than those without fine-tuning. To address these limitations, we propose Weakly-supervised Partial Contrastive Learning (WPCL), a method that enhances an agent’s ability to identify objects from dynamic viewpoints in VLN scenarios by effectively integrating pre-trained VLM knowledge into the perception process, without requiring VLM fine-tuning. Our method enhances the agent’s ability to interpret and respond to environmental cues while ensuring computational efficiency. Experimental results have shown that our method outperforms the baseline methods on multiple benchmarks, which validates the effectiveness, robustness, and generalizability of our method.

Understanding a driver's behavior and intentions is important for potential risk assessment and early accident prevention. Safety and driver assistance systems can be tailored to individual drivers' behavior, significantly enhancing their effectiveness. However, existing datasets are limited in describing and explaining general vehicle movements based on external visual evidence. This paper introduces a benchmark, PDB-Eval, for a detailed understanding of Personalized Driver Behavior, and aligning Large Multimodal Models (MLLMs) with driving comprehension and reasoning. Our benchmark consists of two main components, PDB-X and PDBQA. PDB-X can evaluate MLLMs' understanding of temporal driving scenes. Our dataset is designed to find valid visual evidence from the external view to explain the driver's behavior from the internal view. To align MLLMs' reasoning abilities with driving tasks, we propose PDB-QA as a visual explanation question-answering task for MLLM instruction fine-tuning. As a generic learning task for generative models like MLLMs, PDB-QA can bridge the domain gap without harming MLLMs' generalizability. Our evaluation indicates that fine-tuning MLLMs on fine-grained descriptions and explanations can effectively bridge the gap between MLLMs and the driving domain, which improves zero-shot performance on question-answering tasks by up to 73.2%. We further evaluate the MLLMs fine-tuned on PDB-X in Brain4Cars' intention prediction and AIDE's recognition tasks. We observe up to 12.5% performance improvements on the turn intention prediction task in Brain4Cars, and consistent performance improvements up to 11.0% on all tasks in AIDE.

Recommender system (RecSys) plays important roles in helping users navigate, discover, and consume massive and highly-dynamic information. Today, many RecSys solutions deployed in the real world rely on categorical user profiles and/or pre-calculated recommendation actions that stay static during a user session. However, recent trends suggest that RecSys need to model user intent in real time and constantly adapt to meet user needs at the moment or change user behavior in situ. There are three primary drivers for this emerging need of online adaptation. First, in order to meet the increasing demand for a better personalized experience, the personalization dimensions and space will grow larger and larger. It would not be feasible to pre-compute recommended actions for all personalization scenarios beyond a certain scale. Second, in many settings the system does not have user prior history to leverage. Estimating user intent in real time is the only way to personalize. As various consumer privacy laws tighten, it is foreseeable that many businesses will reduce their reliance on static user profiles. Therefore, it makes the modeling of user intent in real time an important research topic. Third, a user's intent often changes within a session and between sessions, and user behavior could shift significantly during dramatic events. Therefore, it is important to investigate more on online and adaptive recommender systems (OARS) that can adapt in real time to meet user needs and be robust against distribution shifts. Every year, the organizers survey the most important topics for OARS and propose a new workshop program. In light of the recent advancement of (multi-modal) LLMs in RecSys, in this new edition, we decide to formally add the new topic of (multi-modal) LLM models in OARS. We will invite experts and papers in the field to disseminate new knowledge and foster further advancements.

Semantic ID-based recommendation models tokenize each item into a small number of discrete tokens that preserve specific semantics, leading to better performance, scalability, and memory efficiency. While recent models adopt a generative approach, they often suffer from inefficient inference due to the reliance on resource-intensive beam search and multiple forward passes through the neural sequence model. As a result, the length of semantic IDs is typically restricted (e.g., to just 4 tokens), limiting their expressiveness. To address these challenges, we propose RPG, a lightweight framework for semantic ID-based recommendation. The key idea is to produce unordered, long semantic IDs, allowing the model to predict all tokens in parallel. We train the model to predict each token independently using a multi-token prediction loss, directly integrating semantics into the learning objective. During inference, we construct a graph connecting similar semantic IDs and guide decoding to avoid generating invalid IDs. Experiments show that scaling up semantic ID length to 64 enables RPG to outperform generative baselines by an average of 12.6% on the NDCG@10, while also improving inference efficiency. Code is available at: https://github.com/facebookresearch/RPG_KDD2025.

Conversational recommender systems engage users in dialogues to refine their needs and provide more personalized suggestions. Although textual information suffices for many domains, visually driven categories such as fashion or home decor potentially require detailed visual information related to color, style, or design. To address this challenge, we propose LaViC (Large Vision-Language Conversational Recommendation Framework), a novel approach that integrates compact image representations into dialogue-based recommendation systems. LaViC leverages a large vision-language model in a two-stage process: (1) visual knowledge self-distillation, which condenses product images from thousands of tokens into a small set of visual tokens in a self-distillation manner, significantly reducing computational overhead, and (2) recommendation fine-tuning, which enables the model to incorporate both dialogue context and distilled visual tokens, providing a unified mechanism for capturing textual and visual features. To support rigorous evaluation of visually-aware conversational recommendation, we construct a new dataset by aligning Reddit conversations with Amazon product listings across multiple visually oriented categories (e.g., fashion, beauty, and home). This dataset covers realistic user queries and product appearances in domains where visual details are crucial. Extensive experiments demonstrate that LaViC significantly outperforms text-only conversational recommendation methods and open-source vision-language baselines. Moreover, LaViC achieves competitive or superior accuracy compared to prominent proprietary baselines (e.g., GPT-3.5-turbo, GPT-4o-mini, and GPT-4o), demonstrating the necessity of explicitly using visual data for capturing product attributes and showing the effectiveness of our vision-language integration. Our code and dataset are available at https://github.com/jeon185/LaViC.

Large Language Models (LLMs) have recently emerged as promising tools for recommendation thanks to their advanced textual understanding ability and context-awareness. Despite the current practice of training and evaluating LLM-based recommendation (LLM4Rec) models under a sequential recommendation scenario, we found that whether these models understand the sequential information inherent in users' item interaction sequences has been largely overlooked. In this paper, we first demonstrate through a series of experiments that existing LLM4Rec models do not fully capture sequential information both during training and inference. Then, we propose a simple yet effective LLM-based sequential recommender, called LLM-SRec, a method that enhances the integration of sequential information into LLMs by distilling the user representations extracted from a pre-trained CF-SRec model into LLMs. Our extensive experiments show that LLM-SRec enhances LLMs' ability to understand users' item interaction sequences, ultimately leading to improved recommendation performance. Furthermore, unlike existing LLM4Rec models that require fine-tuning of LLMs, LLM-SRec achieves state-of-the-art performance by training only a few lightweight MLPs, highlighting its practicality in real-world applications. Our code is available at https://github.com/Sein-Kim/LLM-SRec.

Building personalized recommender systems is a cornerstone of the modern data mining and applied machine learning (ML) community. Modern online platforms have a confluence of data including user-item interaction graphs, user and item-associated semantics (text, visual content, etc.), and metadata. Recent advancements in generative models and semantic encoders via large language models (LLMs), visual and audio encoders have significantly impacted research in relevant domains, enabling new directions in knowledge discovery and ability of models to better incorporate semantic context. This workshop bridges the research gap between the use of generative models and recommendation for personalized systems. We will focus on topics spanning the interplay between such models and conventional personalized systems.

The field of information retrieval has significantly transformed with the integration of AI technologies. AI agents, especially those leveraging LLMs and vast computational power, have revolutionized in- formation retrieval, processing, and presentation. LLM agents, with advanced memory, reasoning, and planning capabilities, can perform complex tasks, engage in coherent conversations, and provide personalized responses. Despite these advancements, challenges such as ensuring relevance and accuracy, mitigating biases, providing real-time responses, and maintaining data security remain. This workshop aims to explore these challenges, share innovative solutions, and discuss future directions. It will provide a platform to bring together researchers and practitioners to discuss the latest theoretical advancements and practical implementations of AI agents in information retrieval. Topics include AI in search, recommendation, and personalization systems. By gathering a diverse group of experts, the workshop seeks to deepen the understanding of AI agents in information retrieval, advance the field, and enhance its societal impact. Participants will gain insights into cutting-edge research and emerging trends, and foster knowledge exchange and collaboration within the community.

Large multimodal models (LMMs) exhibit enhanced capabilities in understanding and generating both textual and visual content. By leveraging item metadata, LMMs are also applied for recommendation and demonstrate improvements across diverse scenarios. However, the majority of existing methods explore static item attributes without considering additional contextual information (e.g., price, brand). Moreover, overlooking the interaction between the retrieval and ranking stages may lead to suboptimal solutions for fine-grained recommendations. In this work, we introduce PRIME: preference-optimized retrieval and ranking for efficient multimodal recommendation. PRIME operates in two stages: (i) a lightweight retriever identifies potential candidate items; (ii) an LMM learns to rank the retrieved candidates with detailed user history and multimodal features (e.g., text and image attributes). These features are incorporated into a carefully designed prompt, facilitating fine-grained transition patterns for user preference understanding. To optimize the inference efficiency of PRIME, we introduce verbalizer-based inference, which computes ranking scores for all candidate items in a single forward pass. Furthermore, we employ the LMM ranker to provide feedback on sampled candidate sets, enabling online preference optimization that refines the retriever model and improves the alignment between retrieval and ranking. As a result, PRIME can capture subtle user intentions and efficiently rank candidate items with minimal inference costs. Extensive experiments show the effectiveness and efficiency of PRIME, which consistently achieves superior performance over baseline methods.

Tingchen Fu, Yupeng Hou, Julian McAuley, Rui Yan. 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.

Yu Xia, Junda Wu, Sungchul Kim, Tong Yu, Ryan A. Rossi, Haoliang Wang, Julian McAuley. 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.

Zhouhang Xie, Tushar Khot, Bhavana Dalvi Mishra, Harshit Surana, Julian McAuley, Peter Clark, Bodhisattwa Prasad Majumder. 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.

Forecasting multi-step user behavior trajectories requires reasoning over structured preferences across future actions, a challenge overlooked by traditional sequential recommendation. This problem is critical for applications such as personalized commerce and adaptive content delivery, where anticipating a user’s complete action sequence enhances both satisfaction and business outcomes. We identify an essential limitation of existing paradigms: their inability to capture global, listwise dependencies among sequence items. To address this, we formulate User Behavior Trajectory Prediction (UBTP) as a new task setting that explicitly models longterm user preferences. We introduce Listwise Preference Diffusion Optimization (LPDO), a diffusion-based training framework that directly optimizes structured preferences over entire item sequences. LPDO incorporates a Plackett–Luce supervision signal and derives a tight variational lower bound aligned with listwise ranking likelihoods, enabling coherent preference generation across denoising steps and overcoming the independent-token assumption of prior diffusion methods. To rigorously evaluate multi-step prediction quality, we propose the task-specific metric: Sequential Match (SeqMatch), which measures exact trajectory agreement, and adopt Perplexity (PPL), which assesses probabilistic fidelity. Extensive experiments on real-world user behavior benchmarks demonstrate that LPDO consistently outperforms state-of-the-art baselines, establishing a new benchmark for structured preference learning with diffusion models.

Short videos are an effective tool for promoting contents and improving knowledge accessibility. While existing extractive video summarization methods struggle to produce a coherent narrative, existing abstractive methods cannot `quote' from the input videos, i.e., inserting short video clips in their outputs. In this work, we explore novel video editing models for generating shorts that feature a coherent narrative with embedded video insertions extracted from a long input video. We propose a novel retrieval-embedded generation framework that allows a large language model to quote multimodal resources while maintaining a coherent narrative. Our proposed REGen system first generates the output story script with quote placeholders using a finetuned large language model, and then uses a novel retrieval model to replace the quote placeholders by selecting a video clip that best supports the narrative from a pool of candidate quotable video clips. We examine the proposed method on the task of documentary teaser generation, where short interview insertions are commonly used to support the narrative of a documentary. Our objective evaluations show that the proposed method can effectively insert short video clips while maintaining a coherent narrative. In a subjective survey, we show that our proposed method outperforms existing abstractive and extractive approaches in terms of coherence, alignment, and realism in teaser generation.

We introduce NAILS (Normative Alignment of Recommender Systems via Internal Label Shift), a simple and scalable method for aligning recommendation outputs with target distributions over item-level attributes, such as categories. Recommender systems optimized solely for user engagement often fail to satisfy broader normative objectives, including fairness, diversity, and editorial values. NAILS modifies the user-conditional item distribution to induce a specified marginal distribution over attributes while preserving the preferences learned by an existing recommender system and requiring no model retraining. We formulate this problem as a form of label shift applied internally within a hierarchical classification framework. By adopting a stakeholder-centric perspective, NAILS enables recommendation outputs to be aligned with global normative objectives. Empirically, we show that NAILS consistently improves attribute-level alignment with minimal impact on user engagement, providing a practical mechanism for value-driven recommendation.

Recommender systems powered by generative models (Gen-RecSys) extend beyond classical item-ranking by producing open-ended content, which simultaneously unlocks richer user experiences and introduces new risks. On one hand, these systems can enhance personalization and appeal through dynamic explanations and multi-turn dialogues. On the other hand, they might venture into unknown territory-hallucinating nonexistent items, amplifying bias, or leaking private information. Traditional accuracy metrics cannot fully capture these challenges, as they fail to measure factual correctness, content safety, or alignment with user intent.

We present GENNEXT, a workshop dedicated to exploring the integration of language agents, generative models, and conversational AI within information retrieval (IR) and recommender systems (RS). Building on the success of our recent RecSys'24 workshop, GENNEXT aims to advance discussions on the applications of language agents powered by Large Language Models (LLMs). The workshop will focus on enhancing interactivity between users and systems through multi-turn dialogues, improving creative content generation, advancing personalization, and enabling multifaceted, context-aware decision-making. For example, a language agent could respond to a query like ''Suggest an eco-friendly food tour for a weekend in my city'' by using a recommendation API to identify eateries specializing in sustainable or organic cuisine and a pollution API to ensure the selected routes have low air pollution levels.

This intermediate-level tutorial, titled "Gen-RecSys", merges both industrial and academic perspectives on recent advances in Generative AI for recommender systems (beyond LLMs). It aims to highlight the transformative role of generative models in modern recommender systems, which have significantly impacted the AI field-particularly with the rise of large language models (LLMs) like ChatGPT-and have contributed to a rapid convergence of the fields of search, data mining, and recommendation. By providing attendees with a modern perspective on GenAI applications in recommendation, the tutorial will emphasize how generative models can drive recommendation by unlocking and interacting with rich data representations, including behavioral, textual, and multi-modal data-knowledge highly transferable across many applications of interest to the WSDM community. Participants will learn about the categorization of generative models in recommender systems based on underlying data modalities: (i) ID-based collaborative models, (ii) text-driven models such as LLMs, and (iii) multi-modal models. Within each category, various deep generative model paradigms (e.g., AR, GAN, diffusion models) will be introduced, along with insights into their application areas. The tutorial will also cover evaluation aspects, including benchmarks, metrics, and assessments of social and ethical impacts and harms. This tutorial presents a condensed version of the industrial and academic work featured in the forthcoming book at FntIR 2024-25, titled "Recommendation with Generative Models [7]," and a shorter version prepared, and presented by the team, see GenRecSys-Survey [6].

Large Language Models (LLMs) are revolutionizing conversational recommender systems (CRS) by effectively indexing item content, understanding complex conversational contexts, and generating relevant item titles. However, the autoregressive nature of LLMs, which outputs item titles as a long sequence of subtokens, hinders the ability to efficiently obtain and control recommendations across the entire item set. This challenge in calculating probabilities over all items limits LLMs' potential, such as (1) limiting control over recommendation popularities and (2) preventing the synergy of marrying LLMs and traditional recommender systems (RecSys).

Recent research on explainable recommendation generally frames the task as a standard text generation problem, and evaluates models simply based on the textual similarity between the predicted and ground-truth explanations. However, this approach fails to consider one crucial aspect of the systems: whether their outputs accurately reflect the users' (post-purchase) sentiments, i.e., whether and why they would like and/or dislike the recommended items. To shed light on this issue, we introduce new datasets and evaluation methods that focus on the users' sentiments. Specifically, we construct the datasets by explicitly extracting users' positive and negative opinions from their post-purchase reviews using an LLM, and propose to evaluate systems based on whether the generated explanations 1) align well with the users' sentiments, and 2) accurately identify both positive and negative opinions of users on the target items. We benchmark several recent models on our datasets and demonstrate that achieving strong performance on existing metrics does not ensure that the generated explanations align well with the users' sentiments. Lastly, we find that existing models can provide more sentiment-aware explanations when the users' (predicted) ratings for the target items are directly fed into the models as input. The datasets and benchmark implementation are available at: https://github.com/jchanxtarov/sent_xrec.

Recent advances in recommender systems have proved the potential of reinforcement learning (RL) to handle the dynamic evolution processes between users and recommender systems. However, learning to train an optimal RL agent is generally impractical with commonly sparse user feedback data in the context of recommender systems. To circumvent the lack of interaction of current RL-based recommender systems, we propose to learn a general model-agnostic counterfactual synthesis (MACS) policy for counterfactual user interaction data augmentation. The counterfactual synthesis policy aims to synthesize counterfactual states while preserving significant information in the original state relevant to the user's interests, building upon two different training approaches we designed: learning with expert demonstrations and joint training. As a result, the synthesis of each counterfactual data is based on the current recommendation agent's interaction with the environment to adapt to users' dynamic interests. We integrate the proposed policy deep deterministic policy gradient (DDPG), soft actor critic (SAC), and twin delayed DDPG (TD3) in an adaptive pipeline with a recommendation agent that can generate counterfactual data to improve the performance of recommendation. The empirical results on both online simulation and offline datasets demonstrate the effectiveness and generalization of our counterfactual synthesis policy and verify that it improves the performance of RL recommendation agents.

Conversational critiquing in recommender systems offers a way for users to engage in multi-turn conversations to find items they enjoy. For users to trust an agent and give effective feedback, the recommender system must be able to explain its suggestions and rationales. We develop a two-part framework for training multi-turn conversational critiquing in recommender systems that provide recommendation rationales that users can effectively interact with to receive better recommendations. First, we train a recommender system to jointly suggest items and explain its reasoning via subjective rationales. We then fine-tune this model to incorporate iterative user feedback via self-supervised bot-play. Experiments on three real-world datasets demonstrate that our system can be applied to different recommendation models across diverse domains to achieve state-of-the-art performance in multi-turn recommendation. Human studies show that systems trained with our framework provide more useful, helpful, and knowledgeable suggestions in warm- and cold-start settings. Our framework allows us to use only product reviews during training, avoiding the need for expensive dialog transcript datasets that limit the applicability of previous conversational recommender agents.

Recommender systems play a vital role in shaping user experiences across domains such as e-commerce, media, and social platforms. However, challenges such as data scarcity, domain isolation, and the need for greater interpretability continue to hinder their progress. Knowledge transfer has emerged as a transformative approach to address these limitations by leveraging information across domains, models, tasks, and modalities. This special issue explores the role of knowledge transfer in advancing recommender systems and highlights innovative research contributions. Featured works address a variety of challenges, including cold-start problems, cross-domain recommendations, knowledge graph integration, federated learning, explainability, and so on. These studies open discussion toward next-generation recommender systems when transferring the knowledge.

Understanding and accurately explaining compatibility relationships between fashion items is a challenging problem in the burgeoning domain of AI-driven outfit recommendations. Present models, while making strides in this area, still occasionally fall short, offering explanations that can be elementary and repetitive. This work aims to address these shortcomings by introducing the Pair Fashion Explanation (PFE) dataset, a unique resource that has been curated to illuminate these compatibility relationships. Furthermore, we propose an innovative two stage pipeline model that leverages this dataset. This fine-tuning allows the model to generate explanations that convey the compatibility relationships between items. Our experiments showcase the model's potential in crafting descriptions that are knowledgeable, aligned with ground-truth matching correlations, and that produce understandable and informative descriptions, as assessed by both automatic metrics and human evaluation. Our code and data are released at https://github.com/wangyu-ustc/PairFashionExplanation.

We present a new Python toolkit called RecWizard for Conversational Recommender Systems (CRS). RecWizard offers support for development of models and interactive user interface, drawing from the best practices of the Huggingface ecosystems. CRS with RecWizard are modular, portable, interactive and Large Language Models (LLMs)-friendly, to streamline the learning process and reduce the additional effort for CRS research. For more comprehensive information about RecWizard, please check our GitHub https://github.com/McAuley-Lab/RecWizard.

Junda Wu, Tong Yu, Xiang Chen, Haoliang Wang, Ryan Rossi, Sungchul Kim, Anup Rao, Julian McAuley. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

While sequential recommendation achieves significant progress on capturing user-item transition patterns, transferring such large-scale recommender systems remains challenging due to the disjoint user and item groups across domains. In this paper, we propose a vector quantized meta learning for transferable sequential recommenders (MetaRec). Without requiring additional modalities or shared information across domains, our approach leverages user-item interactions from multiple source domains to improve the target domain performance. To solve the input heterogeneity issue, we adopt vector quantization that maps item embeddings from heterogeneous input spaces to a shared feature space. Moreover, our meta transfer paradigm exploits limited target data to guide the transfer of source domain knowledge to the target domain (i.e., learn to transfer). In addition, MetaRec adaptively transfers from multiple source tasks by rescaling meta gradients based on the source-target domain similarity, enabling selective learning to improve recommendation performance. To validate the effectiveness of our approach, we perform extensive experiments on benchmark datasets, where MetaRec consistently outperforms baseline methods by a considerable margin.

Recommender system (RecSys) plays important roles in helping users navigate, discover, and consume massive and highly-dynamic information. Today, many RecSys solutions deployed in the real world rely on categorical user-profiles and/or pre-calculated recommendation actions that stay static during a user session. However, recent trends suggest that RecSys need to model user intent in real time and constantly adapt to meet user needs at the moment or change user behavior in situ. There are three primary drivers for this emerging need of online adaptation. First, in order to meet the increasing demand for a better personalized experience, the personalization dimensions and space will grow larger and larger. It would not be feasible to pre-compute recommended actions for all personalization scenarios beyond a certain scale. Second, in many settings the system does not have user prior history to leverage. Estimating user intent in real time is the only feasible way to personalize. As various consumer privacy laws tighten, it is foreseeable that many businesses will reduce their reliance on static user profiles. Therefore, it makes the modeling of user intent in real time an important research topic. Third, a user's intent often changes within a session and between sessions, and user behavior could shift significantly during dramatic events. Therefore, it is important to investigate more on online and adaptive recommender system (OARS) that can adapt in real time to meet user needs and be robust against distribution shifts. Every year, the organizers survey the most important topics for OARS and propose a new workshop program. In light of the recent advancement of LLMs and foundation models in RecSys, in this new edition, we decide to formally add the new topic of foundation and LLM models in OARS. We will invite experts and papers in the field to facilitate its further advancement. Our workshop offers a focused discussion of the new study and application of OARS, and will bring together an interdisciplinary community of researchers and practitioners from both industry and academia to discuss on new topics in the area, grow a community, and push the direction forward.

Calibrated recommendation, which aims to maintain personalized proportions of categories within recommendations, is crucial in practical scenarios since it enhances user satisfaction by reflecting diverse interests. However, achieving calibration in a sequential setting (i.e., calibrated sequential recommendation) is challenging due to the need to adapt to users' evolving preferences. Previous methods typically leverage reranking algorithms to calibrate recommendations after training a model without considering the effect of calibration and do not effectively tackle the conflict between relevance and calibration during the reranking process. In this work, we propose LeapRec (Calibration-Disentangled Learning and Relevance-Prioritized Reranking), a novel approach for the calibrated sequential recommendation that addresses these challenges. LeapRec consists of two phases, model training phase and reranking phase. In the training phase, a backbone model is trained using our proposed calibration-disentangled learning-to-rank loss, which optimizes personalized rankings while integrating calibration considerations. In the reranking phase, relevant items are prioritized at the top of the list, with items needed for calibration following later to address potential conflicts between relevance and calibration. Through extensive experiments on four real-world datasets, we show that LeapRec consistently outperforms previous methods in the calibrated sequential recommendation. Our code is available at https://github.com/jeon185/LeapRec.

The task of multi-behavioral sequential recommendation (MBSR) has grown in importance in personalized recommender systems, aiming to incorporate behavior types of interactions for better recommendations. Existing approaches focus on the next-item prediction objective, neglecting the value of integrating the target behavior type into the learning objective. In this paper, we propose MBGen, a novel Multi-Behavioral sequential Generative recommendation framework. We model the MBSR task into a consecutive two-step process: (1) given item sequences, MBGen first predicts the next behavior type to frame the user intention, (2) given item sequences and a target behavior type, MBGen then predicts the next items. To model such a two-step process, we tokenize both behaviors and items into tokens and construct one single token sequence with both behaviors and items placed interleaved. Furthermore, we design a unified generative recommendation paradigm that learns to autoregressive generate next behavior and item tokens, naturally enabling a multi-task capability. Additionally, we exploit the heterogeneous nature of token sequences in the generative recommendation and propose a position-routed sparse architecture to efficiently scale up models under the generative recommendation paradigm. Extensive experiments on real-world public datasets demonstrate that MBGen significantly outperforms existing MBSR models across multiple tasks.

Certain abilities of Transformer-based language models consistently emerge in their later layers.Previous research has leveraged this phenomenon to improve factual accuracy through self-contrast, penalizing early-exit predictions based on the premise that later-layer updates are more factually reliable than earlierlayer associations.We observe a similar pattern for fine-grained emotion classification in text, demonstrating that self-contrast can enhance encoder-based text classifiers.Additionally, we reinterpret self-contrast as a form of linear extrapolation, which motivates a refined approach that dynamically adjusts the contrastive strength based on the selected intermediate layer.Experiments across multiple models and emotion classification datasets show that our method outperforms standard classification techniques in fine-grained emotion classification tasks.

Large Language Models (LLMs) have greatly advanced code auto-completion systems, with a potential for substantial productivity enhancements for developers. However, current benchmarks mainly focus on single-file tasks, leaving an assessment gap for more complex, real-world, multi-file programming scenarios. To fill this gap, we introduce RepoBench, a new benchmark specifically designed for evaluating repository-level code auto-completion systems. RepoBench consists of three interconnected evaluation tasks: RepoBench-R (Retrieval), RepoBench-C (Code Completion), and RepoBench-P (Pipeline). Each task respectively measures the system's ability to retrieve the most relevant code snippets from other files as cross-file context, predict the next line of code with cross-file and in-file context, and handle complex tasks that require a combination of both retrieval and next-line prediction. RepoBench aims to facilitate a more complete comparison of performance and encouraging continuous improvement in auto-completion systems. RepoBench is actively maintained with the latest code, serving as a live benchmark publicly available at https://github.com/Leolty/repobench.

The vast applications of deep generative models are anchored in three core capabilities---*generating* new instances, *reconstructing* inputs, and learning compact *representations*---across various data types, such as discrete text/protein sequences and continuous images. Existing model families, like variational autoencoders (VAEs), generative adversarial networks (GANs), autoregressive models, and (latent) diffusion models, generally excel in specific capabilities and data types but fall short in others. We introduce *Generalized* ***E****ncoding*-***D****ecoding ****D****iffusion ****P****robabilistic ****M****odels* (EDDPMs) which integrate the core capabilities for broad applicability and enhanced performance. EDDPMs generalize the Gaussian noising-denoising in standard diffusion by introducing parameterized encoding-decoding. Crucially, EDDPMs are compatible with the well-established diffusion model objective and training recipes, allowing effective learning of the encoder-decoder parameters *jointly* with diffusion. By choosing appropriate encoder/decoder (e.g., large language models), EDDPMs naturally apply to different data types. Extensive experiments on text, proteins, and images demonstrate the flexibility to handle diverse data and tasks and the strong improvement over various existing models. Code is available at https://github.com/guangyliu/EDDPM .

We propose SelfVC, a training strategy to iteratively improve a voice conversion model with self-synthesized examples. Previous efforts on voice conversion focus on factorizing speech into explicitly disentangled representations that separately encode speaker characteristics and linguistic content. However, disentangling speech representations to capture such attributes using task-specific loss terms can lead to information loss. In this work, instead of explicitly disentangling attributes with loss terms, we present a framework to train a controllable voice conversion model on entangled speech representations derived from self-supervised learning (SSL) and speaker verification models. First, we develop techniques to derive prosodic information from the audio signal and SSL representations to train predictive submodules in the synthesis model. Next, we propose a training strategy to iteratively improve the synthesis model for voice conversion, by creating a challenging training objective using self-synthesized examples. We demonstrate that incorporating such self-synthesized examples during training improves the speaker similarity of generated speech as compared to a baseline voice conversion model trained solely on heuristically perturbed inputs. Our framework is trained without any text and achieves state-of-the-art results in zero-shot voice conversion on metrics evaluating naturalness, speaker similarity, and intelligibility of synthesized audio.

We propose Diffusion Inference-Time T-Optimization (DITTO), a general-purpose framework for controlling pre-trained text-to-music diffusion models at inference-time via optimizing initial noise latents. Our method can be used to optimize through any differentiable feature matching loss to achieve a target (stylized) output and leverages gradient checkpointing for memory efficiency. We demonstrate a surprisingly wide-range of applications for music generation including inpainting, outpainting, and looping as well as intensity, melody, and musical structure control – all without ever fine-tuning the underlying model. When we compare our approach against related training, guidance, and optimization-based methods, we find DITTO achieves state-of-the-art performance on nearly all tasks, including outperforming comparable approaches on controllability, audio quality, and computational efficiency, thus opening the door for high-quality, flexible, training-free control of diffusion models. Sound examples can be found at https://ditto-music.github.io/web/.

Existing Large Language Models (LLMs) usually remain static after deployment, which might make it hard to inject new knowledge into the model. We aim to build models containing a considerable portion of self-updatable parameters, enabling the model to integrate new knowledge effectively and efficiently. To this end, we introduce MEMORYLLM, a model that comprises a transformer and a fixed-size memory pool within the latent space of the transformer. MEMORYLLM can self-update with text knowledge and memorize the knowledge injected earlier. Our evaluations demonstrate the ability of MEMORYLLM to effectively incorporate new knowledge, as evidenced by its performance on model editing benchmarks. Meanwhile, the model exhibits long-term information retention capacity, which is validated through our custom-designed evaluations and long-context benchmarks. MEMORYLLM also shows operational integrity without any sign of performance degradation even after nearly a million memory updates. Our code and model are open-sourced at https://github.com/wangyu-ustc/MemoryLLM.

Traditional recommender systems typically use user-item rating histories as their main data source. However, deep generative models now have the capability to model and sample from complex data distributions, including user-item interactions, text, images, and videos, enabling novel recommendation tasks. This comprehensive, multidisciplinary survey connects key advancements in RS using Generative Models (Gen-RecSys), covering: interaction-driven generative models; the use of large language models (LLM) and textual data for natural language recommendation; and the integration of multimodal models for generating and processing images/videos in RS. Our work highlights necessary paradigms for evaluating the impact and harm of Gen-RecSys and identifies open challenges. This survey accompanies a "tutorial" presented at ACM KDD'24, with supporting materials provided at: https://encr.pw/vDhLq.

Personalization is key in understanding user behavior and has been a main focus in the fields of knowledge discovery and information retrieval. Building personalized recommender systems is especially important now due to the vast amount of user-generated textual content, which offers deep insights into user preferences. The recent advancements in Large Language Models (LLMs) have significantly impacted research areas, mainly in Natural Language Processing and Knowledge Discovery, giving these models the ability to handle complex tasks and learn context. However, the use of generative models and user-generated text for personalized systems and recommendation is relatively new and has shown some promising results. This workshop is designed to bridge the research gap in these fields and explore personalized applications and recommender systems. We aim to fully leverage generative models to develop AI systems that are not only accurate but also focused on meeting individual user needs. Building upon the momentum of previous successful forums, this workshop seeks to engage a diverse audience from academia and industry, fostering a dialogue that incorporates fresh insights and anticipates over 50 attendees, including key stakeholders in the field.

The long-tail recommendation is a challenging task for traditional recommender systems, due to data sparsity and data imbalance issues. The recent development of large language models (LLMs) has shown their abilities in complex reasoning, which can help to deduce users' preferences based on very few previous interactions. However, since most LLM-based systems rely on items' semantic meaning as the sole evidence for reasoning, the collaborative information of user-item interactions is neglected, which can cause the LLM's reasoning to be misaligned with task-specific collaborative information of the dataset. To further align LLMs' reasoning to task-specific user-item interaction knowledge, we introduce collaborative retrieval-augmented LLMs, CoRAL, which directly incorporate collaborative evidence into the prompts. Based on the retrieved user-item interactions, the LLM can analyze shared and distinct preferences among users, and summarize the patterns indicating which types of users would be attracted by certain items. The retrieved collaborative evidence prompts the LLM to align its reasoning with the user-item interaction patterns in the dataset. However, since the capacity of the input prompt is limited, finding the minimally-sufficient collaborative information for recommendation tasks can be challenging. We propose to find the optimal interaction set through a sequential decision-making process and develop a retrieval policy learned through a reinforcement learning (RL) framework, CoRAL. Our experimental results show that CoRAL can significantly improve LLMs' reasoning abilities on specific recommendation tasks. Our analysis also reveals that CoRAL can more efficiently explore collaborative information through reinforcement learning.

Yu Xia, Tong Yu, Zhankui He, Handong Zhao, Julian McAuley, Shuai Li. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Se-eun Yoon, Zhankui He, Jessica Echterhoff, Julian McAuley. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

We present an overview of a workshop focused on the exploration of generative models within recommender systems (RS). It highlights the dual nature of these technologies: on the one hand, they offer groundbreaking opportunities for enhancing RS through improved personalization, innovative content creation, and interactive user experiences; on the other hand, they introduce a range of challenges, including bias, misinformation, privacy concerns, and environmental impact.

Inferring user intent in recommender systems can help performance but is difficult because intent is personal and not directly observable. Previous work has leveraged signals to stand as a proxy for intent (e.g. user interactions with resource pages), but such signals are not always available. In this paper, we instead recognize that certain events, which are observable, directly influence user intent. For example, after a flood, home improvement customers are more likely to undertake a renovation project to dry out their basement. We introduce MAWI Rec, a recommender system that leverages severe weather data to improve recommendation. Our weather-aware system achieves a significant improvement over a state-of-the-art baseline for online and in-store datasets of home improvement customers. This gain is most significant for weather-related product categories such as roof panels and flashings.

Conversational recommender systems (CRS) should understand users’ expressed interests, which are frequently semantically rich and knowledge-intensive. Prior works attempt to address this challenge by using external knowledge bases or parametric knowledge in large language models (LLMs). In this paper, we study a complementary solution, exploiting item knowledge in the training data. We hypothesize that many inference-time user requests can be answered by reusing popular crowd-written answers associated with similar training queries. Following this intuition, we define a class of neighborhood-based CRS that makes recommendations by identifying items commonly associated with similar training dialogue contexts. Experiments on Inspired, Redial, and Reddit-Movie benchmarks show our method outperforms state-of-the-art LLMs with 2 billion parameters, and offers on-par performance to 7 billion parameter models while using over 170 times less GPU memory. We also show neighborhood and model-based predictions can be combined to achieve further performance improvements1.

Much existing literature on fair recommendation (i.e., group fairness) leverages users' demographic attributes (e.g., gender) to develop fair recommendation methods. However, in real-world scenarios, due to privacy concerns and convenience considerations, users may not be willing to share their demographic information with the system, which limits the application of many existing methods. Moreover, sequential recommendation (SR) models achieve state-of-the-art performance compared to traditional collaborative filtering (CF) recommenders, and can represent users solely using user-item interactions (user-free). This leaves a wrong impression that SR models are free from group unfairness by design. In this work, we explore a critical question: how can we build a fair sequential recommendation system without even knowing user demographics? To address this problem, we propose Agnostic FairSeqRec (A-FSR): a model-agnostic and demographic-agnostic debiasing framework for sequential recommendation without requiring users' demographic attributes. Firstly, A-FSR reduces the correlation between the potential stereotypical patterns in the input sequences and final recommendations via Dirichlet neighbor smoothing. Secondly, A-FSR estimates an under-represented group of sequences via a gradient-based heuristic, and implicitly moves training focus towards the under-represented group by minimizing a distributionally robust optimization (DRO) based objective. Results on real-world datasets show that A-FSR achieves significant improvements on group fairness in sequential recommendation, while outperforming other state-of-the-art baselines.

During online browsing, e.g. when looking to select a movie to watch, we are often confronted with multiple rejection-selection steps which can lead to tens or hundreds of decisions made in quick succession. It is unclear if showing the next “best” item, as often employed by standard recommenders, is the most efficient way to help users select an item. In this work, we show that we can reduce the number of decisions to selection with a reinforcement learning-based Decision Minimizer Network (DMN). By implementing a step-aware reward function we can penalize long sequences, leading to fewer decisions having to be made by humans. Using a task to select a movie to watch, we show that we can reduce the number of decisions to selection by 39% compared to heuristic strategies and by 20% compared to standard recommender while increasing user selection satisfaction. Minimizing the number of decision steps can finally help to reduce decision fatigue, which refers to the deteriorating quality of decisions made by an individual after a long session of decision steps, and help to prevent infinite scrolling.

Concept bottleneck models have been successfully used for explainable machine learning by encoding information within the model with a set of human-defined concepts. In the context of human-assisted or autonomous driving, explainability models can help user acceptance and understanding of decisions made by the autonomous vehicle, which can be used to rationalize and explain driver or vehicle behavior. We propose a new approach using concept bottlenecks as visual features for control command predictions and explanations of user and vehicle behavior. We learn a human-understandable concept layer that we use to explain sequential driving scenes while learning vehicle control commands. This approach can then be used to determine whether a change in a preferred gap or steering commands from a human (or autonomous vehicle) is led by an external stimulus or change in preferences. We achieve competitive performance to latent visual features while gaining interpretability within our model setup.1

State-of-the-art sequential recommendation relies heavily on self-attention-based recommender models. Yet such models are computationally expensive and often too slow for real-time recommendation. Furthermore, the self-attention operation is performed at a sequence-level, thereby making low-cost incremental inference challenging. Inspired by recent advances in efficient language modeling, we propose linear recurrent units for sequential recommendation (LRURec). Similar to recurrent neural networks, LRURec offers rapid inference and can achieve incremental inference on sequential inputs. By decomposing the linear recurrence operation and designing recursive parallelization in our framework, LRURec provides the additional benefits of reduced model size and parallelizable training. Moreover, we optimize the architecture of LRURec by implementing a series of modifications to address the lack of non-linearity and improve training dynamics. To validate the effectiveness of our proposed LRURec, we conduct extensive experiments on multiple real-world datasets and compare its performance against state-of-the-art sequential recommenders. Experimental results demonstrate the effectiveness of LRURec, which consistently outperforms baselines by a significant margin. Results also highlight the efficiency of LRURec with our parallelized training paradigm and fast inference on long sequences, showing its potential to further enhance user experience in sequential recommendation.

Ranking items regarding individual user interests is a core technique of multiple downstream tasks such as recommender systems. Learning such a personalized ranker typically relies on the implicit feedback from users' past click-through behaviors. However, collected feedback is biased toward previously highly-ranked items and directly learning from it would result in "rich-get-richer" phenomena. In this paper, we propose a simple yet sufficient unbiased learning-to-rank paradigm named InfoRank that aims to simultaneously address both position and popularity biases. We begin by consolidating the impacts of those biases into a single observation factor, thereby providing a unified approach to addressing bias-related issues. Subsequently, we minimize the mutual information between the observation estimation and the relevance estimation conditioned on the input features. By doing so, our relevance estimation can be proved to be free of bias. To implement InfoRank, we first incorporate an attention mechanism to capture latent correlations within user-item features, thereby generating estimations of observation and relevance. We then introduce a regularization term, grounded in conditional mutual information, to promote conditional independence between relevance estimation and observation estimation. Experimental evaluations conducted across three extensive recommendation and search datasets reveal that InfoRank learns more precise and unbiased ranking strategies.

introduction Free Access Share on Just AcceptedSpecial Issue on Responsible Recommender Systems Part 1 Authors: Lina Yao CSIRO's Data61 and University of New South Wales, Australia CSIRO's Data61 and University of New South Wales, Australia 0000-0002-4149-839XSearch about this author , Julian McAuley University of California, USA University of California, USASearch about this author , Xianzhi Wang University of Technology Sydney, Australia University of Technology Sydney, Australia 0000-0001-9582-3445Search about this author , Dietmar Jannach University of Klagenfurt, Austria University of Klagenfurt, Austria 0000-0002-4698-8507Search about this author Authors Info & Claims ACM Transactions on Intelligent Systems and TechnologyAccepted on April 2024https://doi.org/10.1145/3663528Online AM:15 June 2024Publication History 0citation5DownloadsMetricsTotal Citations0Total Downloads5Last 12 Months5Last 6 weeks5 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Publisher SiteeReaderPDF

Modern recommender systems may output considerably different recommendations due to small perturbations in the training data. Changes in the data from a single user will alter the recommendations as well as the recommendations of other users. In applications like healthcare, housing, and finance, this sensitivity can have adverse effects on user experience. We propose a method to stabilize a given recommender system against such perturbations. This is a challenging task due to (1) the lack of a “reference” rank list that can be used to anchor the outputs; and (2) the computational challenges in ensuring the stability of rank lists with respect to all possible perturbations of training data. Our method, FINEST , overcomes these challenges by obtaining reference rank lists from a given recommendation model and then fine-tuning the model under simulated perturbation scenarios with rank-preserving regularization on sampled items. Our experiments on real-world datasets demonstrate that FINEST can ensure that recommender models output stable recommendations under a wide range of different perturbations without compromising next-item prediction accuracy.

Current approaches to zero-shot learning (ZSL) struggle to learn generalizable semantic knowledge capable of capturing complex correlations. Inspired by Spiral Curriculum, which enhances learning processes by revisiting knowledge, we propose a form of spiral learning that revisits visual representations based on a sequence of attribute groups (e.g., a combined group of color and shape). Spiral learning aims to learn generalized local correlations, enabling models to gradually enhance global learning and, thus, understand complex correlations. Our implementation is based on a two-stage reinforced self-revised (RSR) framework: preview and review. RSR first previews visual information to construct diverse attribute groups in a weakly supervised manner. Then, it spirally learns refined localities based on attribute groups and uses localities to revise global semantic correlations. Our framework outperforms state-of-the-art algorithms on four benchmark datasets in both zero-shot and generalized zero-shot settings, which demonstrates the effectiveness of spiral learning in learning generalizable and complex correlations. We also conduct extensive analysis to show that attribute groups and reinforced decision processes can capture complementary semantic information to improve predictions and aid explainability.

Reinforcement learning serves as a potent tool for modeling dynamic user interests within recommender systems, garnering increasing research attention of late. However, a significant drawback persists: its poor data efficiency, stemming from its interactive nature. The training of reinforcement learning-based recommender systems demands expensive online interactions to amass adequate trajectories, essential for agents to learn user preferences. This inefficiency renders reinforcement learning-based recommender systems a formidable undertaking, necessitating the exploration of potential solutions. Recent strides in offline reinforcement learning present a new perspective. Offline reinforcement learning empowers agents to glean insights from offline datasets and deploy learned policies in online settings. Given that recommender systems possess extensive offline datasets, the framework of offline reinforcement learning aligns seamlessly. Despite being a burgeoning field, works centered on recommender systems utilizing offline reinforcement learning remain limited. This survey aims to introduce and delve into offline reinforcement learning within recommender systems, offering an inclusive review of existing literature in this domain. Furthermore, we strive to underscore prevalent challenges, opportunities, and future pathways, poised to propel research in this evolving field.

Manipulated media is becoming a prominent threat due to the recent advances in realistic image and video synthesis techniques. There have been several attempts at detecting synthetically tampered media using machine learning classifiers. However, such classifiers do not generalize well to black-box image synthesis techniques and have been shown to be vulnerable to adversarial examples. To address these challenges, we introduce FaceSigns —a deep learning-based semi-fragile watermarking technique that allows media authentication by verifying an invisible secret message embedded in the image pixels. Instead of identifying and detecting manipulated media using visual artifacts, we propose to proactively embed a semi-fragile watermark into a real image or video so that we can prove its authenticity when needed. FaceSigns is designed to be fragile to malicious manipulations or tampering while being robust to benign operations such as image/video compression, scaling, saturation, contrast adjustments, and so forth. This allows images and videos shared over the internet to retain the verifiable watermark as long as a malicious modification technique is not applied. We demonstrate that our framework can embed a 128-bit secret as an imperceptible image watermark that can be recovered with a high bit recovery accuracy at several compression levels, while being non-recoverable when unseen malicious manipulations are applied. For a set of unseen benign and malicious manipulations studied in our work, our framework can reliably detect manipulated content with an AUC score of 0.996, which is significantly higher than prior image watermarking and steganography techniques.

Sharing medical reports is essential for patient-centered care. A recent line of work has focused on automatically generating reports with NLP methods. However, different audiences have different purposes when writing/reading medical reports – for example, healthcare professionals care more about pathology, whereas patients are more concerned with the diagnosis ("Is there any abnormality?"). The expectation gap results in a common situation where patients find their medical reports to be ambiguous and therefore unsure about the next steps. In this work, we explore the audience expectation gap in healthcare and summarize common ambiguities that lead patients to be confused about their diagnosis into three categories: medical jargon, contradictory findings, and misleading grammatical errors. Based on our analysis, we define a disambiguation rewriting task to regenerate an input to be unambiguous while preserving information about the original content. We further propose a rewriting algorithm based on contrastive pretraining and perturbation-based rewriting. In addition, we create two datasets, OpenI-Annotated based on chest reports and VA-Annotated based on general medical reports, with available binary labels for ambiguity and abnormality presence annotated by radiology specialists. Experimental results on these datasets show that our proposed algorithm effectively rewrites input sentences in a less ambiguous way with high content fidelity. Our code and annotated data will be released to facilitate future research.

Recent models can generate fluent and grammatical synthetic reviews while accurately predicting user ratings. The generated reviews, expressing users' estimated opinions towards related products, are often viewed as natural language ‘rationales’ for the jointly predicted rating. However, previous studies found that existing models often generate repetitive, universally applicable, and generic explanations, resulting in uninformative rationales. Further, our analysis shows that previous models' generated content often contain factual hallucinations. These issues call for novel solutions that could generate both informative and factually grounded explanations. Inspired by recent success in using retrieved content in addition to parametric knowledge for generation, we propose to augment the generator with a personalized retriever, where the retriever's output serves as external knowledge for enhancing the generator. Experiments on Yelp, TripAdvisor, and Amazon Movie Reviews dataset show our model could generate explanations that more reliably entail existing reviews, are more diverse, and are rated more informative by human evaluators.

Despite achieving state-of-the-art performance on many NLP tasks, the high energy cost and long inference delay prevent Transformer-based pretrained language models (PLMs) from seeing broader adoption including for edge and mobile computing. Efficient NLP research aims to comprehensively consider computation, time and carbon emission for the entire life-cycle of NLP, including data preparation, model training and inference. In this survey, we focus on the inference stage and review the current state of model compression and acceleration for pretrained language models, including benchmarks, metrics and methodology.

The “pretraining-and-finetuning” paradigm has become a norm for training domain-specific models in natural language processing and computer vision. In this work, we aim to examine this paradigm for symbolic music generation through leveraging the largest ever symbolic music dataset sourced from the MuseScore forum. We first pretrain a large unconditional transformer model using 1.5 million songs. We then propose a simple technique to equip this pretrained unconditional music transformer model with instrument and genre controls by finetuning the model with additional control tokens. Our proposed representation offers improved high-level controllability and expressiveness against two existing representations. The experimental results show that the proposed model can successfully generate music with user-specified instruments and genre. In a subjective listening test, the proposed model outperforms the pretrained baseline model in terms of coherence, harmony, arrangement and overall quality.

The First Workshop on Personalized Generative AI1 aims to be a cornerstone event fostering innovation and collaboration in the dynamic field of personalized AI. Leveraging the potent capabilities of Large Language Models (LLMs) to enhance user experiences with tailored responses and recommendations, the workshop is designed to address a range of pressing challenges including knowledge gap bridging, hallucination mitigation, and efficiency optimization in handling extensive user profiles. As a nexus for academics and industry professionals, the event promises rich discussions on a plethora of topics such as the development and fine-tuning of foundational models, strategies for multi-modal personalization, and the imperative ethical and privacy considerations in LLM deployment. Through a curated series of keynote speeches, insightful panel discussions, and hands-on sessions, the workshop aspires to be a catalyst in the development of more precise, contextually relevant, and user-centric AI systems. It aims to foster a landscape where generative AI systems are not only responsive but also anticipatory of individual user needs, marking a significant stride in personalized experiences.

In this paper, we present empirical studies on conversational recommendation tasks using representative large language models in a zero-shot setting with three primary contributions. (1) Data: To gain insights into model behavior in "in-the-wild" conversational recommendation scenarios, we construct a new dataset of recommendation-related conversations by scraping a popular discussion website. This is the largest public real-world conversational recommendation dataset to date. (2) Evaluation: On the new dataset and two existing conversational recommendation datasets, we observe that even without fine-tuning, large language models can outperform existing fine-tuned conversational recommendation models. (3) Analysis: We propose various probing tasks to investigate the mechanisms behind the remarkable performance of large language models in conversational recommendation. We analyze both the large language models' behaviors and the characteristics of the datasets, providing a holistic understanding of the models' effectiveness, limitations and suggesting directions for the design of future conversational recommenders.

There has recently been growing interest in the automatic generation of cooking recipes that satisfy some form of dietary restrictions, thanks in part to the availability of online recipe data.Prior studies have used pre-trained language models, or relied on small paired recipe data (e.g., a recipe paired with a similar one that satisfies a dietary constraint).However, pre-trained language models generate inconsistent or incoherent recipes, and paired datasets are not available at scale.We address these deficiencies with RecipeCrit, a hierarchical denoising auto-encoder that edits recipes given ingredientlevel critiques.The model is trained for recipe completion to learn semantic relationships within recipes.Our work's main innovation is our unsupervised critiquing module that allows users to edit recipes by interacting with the predicted ingredients; the system iteratively rewrites recipes to satisfy users' feedback.Experiments on the Recipe1M recipe dataset show that our model can more effectively edit recipes compared to strong language-modeling baselines, creating recipes that satisfy user constraints and are more correct, serendipitous, coherent, and relevant as measured by human judges.

Masked language modeling (MLM) plays a key role in pretraining large language models.But the MLM objective is often dominated by high-frequency words that are sub-optimal for learning factual knowledge.In this work, we propose an approach for influencing MLM pretraining in a way that can improve language model performance on a variety of knowledgeintensive tasks.We force the language model to prioritize informative words in a fully unsupervised way.Experiments demonstrate that the proposed approach can significantly improve the performance of pretrained language models on tasks such as factual recall, question answering, sentiment analysis, and natural language inference in a closed-book setting.

Curated datasets for healthcare are often limited due to the need of human annotations from experts.In this paper, we present MEDEVAL, a multi-level, multi-task, and multi-domain medical benchmark to facilitate the development of language models for healthcare.MEDEVAL is comprehensive and consists of data from several healthcare systems and spans 35 human body regions from 8 examination modalities.With 22,779 collected sentences and 21,228 reports, we provide expert annotations at multiple levels, offering a granular potential usage of the data and supporting a wide range of tasks.Moreover, we systematically evaluated 10 generic and domain-specific language models under zero-shot and finetuning settings, from domain-adapted baselines in healthcare to general-purposed state-of-the-art large language models (e.g., ChatGPT).Our evaluations reveal varying effectiveness of the two categories of language models across different tasks, from which we notice the importance of instruction tuning for few-shot usage of large language models.Our investigation paves the way toward benchmarking language models for healthcare and provides valuable insights into the strengths and limitations of adopting large language models in medical domains, informing their practical applications and future advancements 1 . Chest

Debiasing methods in NLP models traditionally focus on isolating information related to a sensitive attribute (e.g.gender or race).We instead argue that a favorable debiasing method should use sensitive information 'fairly,' with explanations, rather than blindly eliminating it.This fair balance is often subjective and can be challenging to achieve algorithmically.We explore two interactive setups with a frozen predictive model and show that users able to provide feedback can achieve a better and fairer balance between task performance and bias mitigation.In one setup, users, by interacting with test examples, further decreased bias in the explanations (5-8%) while maintaining the same prediction accuracy.In the other setup, human feedback was able to disentangle associated bias and predictive information from the input leading to superior bias mitigation and improved task performance (4-5%) simultaneously.

Engaging with discussion of TV shows online often requires individuals to refrain from consuming show-related content for extended periods to avoid spoilers.While existing research on spoiler detection shows promising results in safeguarding viewers from general spoilers, it fails to address the issue of users abstaining from show-related content during their watch.This is primarily because the definition of a spoiler varies depending on the viewer's progress in the show, and conventional spoiler detection methods lack the granularity to capture this complexity.To tackle this challenge, we propose the task of spoiler matching, which involves assigning an episode number to a spoiler given a specific TV show.We frame this task as semantic text matching and introduce a dataset comprised of comments and episode summaries to evaluate model performance.Given the length of each example, our dataset can also serve as a benchmark for longrange language models. 1 2

Chat models, such as ChatGPT, have shown impressive capabilities and have been rapidly adopted across numerous domains.However, these models are only accessible through a restricted API, creating barriers for new research and progress in the field.We propose a pipeline that can automatically generate a highquality multi-turn chat corpus by leveraging ChatGPT to engage in a conversation with itself.Subsequently, we employ parameter-efficient tuning to enhance LLaMA, an open-source large language model.The resulting model, named Baize, demonstrates good performance in multi-turn dialogues with guardrails that minimize potential risks.Additionally, we propose a new technique called Self-Distill with Feedback, to further improve the performance of the Baize models with feedback from ChatGPT.The Baize models and data are released for research purposes only. 1

Existing approaches for generating multitrack music with transformer models have been limited in terms of the number of instruments, the length of the music segments and slow inference. This is partly due to the memory requirements of the lengthy input sequences necessitated by existing representations. In this work, we propose a new multitrack music representation that allows a diverse set of instruments while keeping a short sequence length. Our proposed Multitrack Music Transformer (MMT) achieves comparable performance with state-of-the-art systems, landing in between two recently proposed models in a subjective listening test, while achieving substantial speedups and memory reductions over both, making the method attractive for real time improvisation or near real time creative applications. Further, we propose a new measure for analyzing musical self-attention and show that the trained model attends more to notes that form a consonant interval with the current note and to notes that are 4N beats away from the current step.

Recent advances in foundation models present new opportunities for interpretable visual recognition – one can first query Large Language Models (LLMs) to obtain a set of attributes that describe each class, then apply vision-language models to classify images via these attributes. Pioneering work shows that querying thousands of attributes can achieve performance competitive with image features. However, our further investigation on 8 datasets reveals that LLM-generated attributes in a large quantity perform almost the same as random words. This surprising finding suggests that significant noise may be present in these attributes. We hypothesize that there exist subsets of attributes that can maintain the classification performance with much smaller sizes, and propose a novel learning-to-search method to discover those concise sets of attributes. As a result, on the CUB dataset, our method achieves performance close to that of massive LLM-generated attributes (e.g., 10k attributes for CUB), yet using only 32 attributes in total to distinguish 200 bird species. Furthermore, our new paradigm demonstrates several additional benefits: higher interpretability and interactivity for humans, and the ability to summarize knowledge for a recognition task.

In this paper, we introduce a new task for code completion that focuses on handling long code input and propose a sparse Transformer model, called LongCoder, to address this task. LongCoder employs a sliding window mechanism for self-attention and introduces two types of globally accessible tokens - bridge tokens and memory tokens - to improve performance and efficiency. Bridge tokens are inserted throughout the input sequence to aggregate local information and facilitate global interaction, while memory tokens are included to highlight important statements that may be invoked later and need to be memorized, such as package imports and definitions of classes, functions, or structures. We conduct experiments on a newly constructed dataset that contains longer code context and the publicly available CodeXGLUE benchmark. Experimental results demonstrate that LongCoder achieves superior performance on code completion tasks compared to previous models while maintaining comparable efficiency in terms of computational resources during inference.

Open vocabulary models (e.g. CLIP) have shown strong performance on zero-shot classification through their ability generate embeddings for each class based on their (natural language) names. Prior work has focused on improving the accuracy of these models through prompt engineering or by incorporating a small amount of labeled downstream data (via finetuning). However, there has been little focus on improving the richness of the class names themselves, which can pose issues when class labels are coarsely-defined and are uninformative. We propose Classification with Hierarchical Label Sets (or CHiLS), an alternative strategy for zero-shot classification specifically designed for datasets with implicit semantic hierarchies. CHiLS proceeds in three steps: (i) for each class, produce a set of subclasses, using either existing label hierarchies or by querying GPT-3; (ii) perform the standard zero-shot CLIP procedure as though these subclasses were the labels of interest; (iii) map the predicted subclass back to its parent to produce the final prediction. Across numerous datasets with underlying hierarchical structure, CHiLS leads to improved accuracy in situations both with and without ground-truth hierarchical information. CHiLS is simple to implement within existing zero-shot pipelines and requires no additional training cost. Code is available at: https://github.com/acmi-lab/CHILS.

Convolutional neural networks (CNNs) are a foun-dational model architecture utilized to perform a wide variety of visual tasks. On image classification tasks CNNs achieve high performance, however model accuracy degrades quickly when inputs are perturbed by distortions such as additive noise or blurring. This drop in performance partly arises from incorrect detection of local features by convolutional layers. In this work, we develop a neuroscience-inspired unsupervised Sleep Replay Consolidation (SRC) algorithm for improving convolutional fil-ter's robustness to perturbations. We demonstrate that sleep-based optimization improves the quality of convolutional layers by the selective modification of spatial gradients across filters. We further show that, compared to other approaches such as fine-tuning, a single sleep phase improves robustness across different types of distortions in a data efficient manner.

Personalized recommender systems fulfill the daily demands of customers and boost online businesses. The goal is to learn a policy that can generate a list of items that matches the user's demand or interest. While most existing methods learn a pointwise scoring model that predicts the ranking score of each individual item, recent research shows that the listwise approach can further improve the recommendation quality by modeling the intra-list correlations of items that are exposed together. This has motivated the recent list reranking and generative recommendation approaches that optimize the overall utility of the entire list. However, it is challenging to explore the combinatorial space of list actions and existing methods that use cross-entropy loss may suffer from low diversity issues. In this work, we aim to learn a policy that can generate sufficiently diverse item lists for users while maintaining high recommendation quality. The proposed solution, GFN4Rec, is a generative method that takes the insight of the flow network to ensure the alignment between list generation probability and its reward. The key advantages of our solution are the log scale reward matching loss that intrinsically improves the generation diversity and the autoregressive item selection model that captures the item mutual influences while capturing future reward of the list. As validation of our method's effectiveness and its superior diversity during active exploration, we conduct experiments on simulated online environments as well as an offline evaluation framework for two real-world datasets.

Recommender systems (RecSys) play important roles in helping users navigate, discover, and consume large and highly dynamic information. Today, many RecSys solutions deployed in the real world rely on categorical user-profiles and/or pre-calculated recommendation actions that stay static during a user session. However, recent trends suggest that RecSys need to model user intent in real time and constantly adapt to meet user needs at the moment or change user behavior in situ. There are three primary drivers for this emerging need of online adaptation. First, in order to meet the increasing demand for a better personalized experience, the personalization dimensions and space will grow larger and larger. It would not be feasible to pre-compute recommended actions for all personalization scenarios beyond a certain scale. Second, in many settings the system does not have user prior history to leverage. Estimating user intent in real time is the only feasible way to personalize. As various consumer privacy laws tighten, it is foreseeable that many businesses will reduce their reliance on static user profiles. Therefore, it makes the modeling of user intent in real time an important research topic. Third, a user's intent often changes within a session and between sessions, and user behavior could shift significantly during dramatic events. A RecSys should adapt in real time to meet user needs and be robust against distribution shifts. The online and adaptive recommender systems (OARS) workshop offers a focused discussion of the study and application of OARS, and will bring together an interdisciplinary community of researchers and practitioners from both industry and academia. KDD, as the premier data science conference, is an ideal venue to gather leaders in the field to further research into OARS and promote its adoption. This workshop is complementary to several sessions of the main conference (e.g., recommendation, reinforcement learning, etc.) and brings them together using a practical and focused application.

Personalized natural language generation for explainable recommendations plays a key role in justifying why a recommendation might match a user's interests. Existing models usually control the generation process by aspect planning. While promising, these aspect-planning methods struggle to generate specific information correctly, which prevents generated explanations from being convincing. In this paper, we claim that introducing lexical constraints can alleviate the above issues. We propose a model, UCEpic, that generates high-quality personalized explanations for recommendation results by unifying aspect planning and lexical constraints in an insertion-based generation manner.

Sequential recommendation aims to model dynamic user behavior from historical interactions. Existing methods rely on either explicit item IDs or general textual features for sequence modeling to understand user preferences. While promising, these approaches still struggle to model cold-start items or transfer knowledge to new datasets. In this paper, we propose to model user preferences and item features as language representations that can be generalized to new items and datasets. To this end, we present a novel framework, named Recformer, which effectively learns language representations for sequential recommendation. Specifically, we propose to formulate an item as a "sentence" (word sequence) by flattening item key-value attributes described by text so that an item sequence for a user becomes a sequence of sentences. For recommendation, Recformer is trained to understand the "sentence" sequence and retrieve the next "sentence". To encode item sequences, we design a bi-directional Transformer similar to the model Longformer but with different embedding layers for sequential recommendation. For effective representation learning, we propose novel pretraining and finetuning methods which combine language understanding and recommendation tasks. Therefore, Recformer can effectively recommend the next item based on language representations. Extensive experiments conducted on six datasets demonstrate the effectiveness of Recformer for sequential recommendation, especially in low-resource and cold-start settings.

Existing explanation models generate only text for recommendations but still struggle to produce diverse contents. In this paper, to further enrich explanations, we propose a new task named personalized showcases, in which we provide both textual and visual information to explain our recommendations. Specifically, we first select a personalized image set that is the most relevant to a user's interest toward a recommended item. Then, natural language explanations are generated accordingly given our selected images. For this new task, we collect a large-scale dataset from Google Maps and construct a high-quality subset for generating multi-modal explanations. We propose a personalized multi-modal framework which can generate diverse and visually-aligned explanations via contrastive learning. Experiments show that our framework benefits from different modalities as inputs, and is able to produce more diverse and expressive explanations compared to previous methods on a variety of evaluation metrics.

The session-based recommendation aims to predict users' immediate next actions based on their short-term behaviors reflected by past and ongoing sessions. Graph neural networks (GNNs) recently dominated the related studies, yet their performance heavily relies on graph structures, which are often predefined, task-specific, and designed heuristically. Furthermore, existing graph-based methods either neglect implicit correlations among items or consider explicit and implicit relationships altogether in the same graphs. We propose to decouple explicit and implicit relationships among items. As such, we can capture the prior knowledge encapsulated in explicit dependencies and learned implicit correlations among items simultaneously in a flexible and more interpretable manner for effective recommendations. We design a dual graph neural network that leverages the feature representations extracted by two GNNs: a graph neural network with a single gate (SG-GNN) and an adaptive graph neural network (A-GNN). The former models explicit dependencies among items. The latter employs a self-learning strategy to capture implicit correlations among items. Our experiments on four real-world datasets show our model outperforms state-of-the-art methods by a large margin, achieving 18.46% and 70.72% improvement in [email protected], and 49.10% and 115.29% improvement in [email protected] on Diginetica and LastFM datasets.

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.

In light of the emergence of deep reinforcement learning (DRL) in recommender systems research and several fruitful results in recent years, this survey aims to provide a timely and comprehensive overview of recent trends of deep reinforcement learning in recommender systems. We start by motivating the application of DRL in recommender systems, followed by a taxonomy of current DRL-based recommender systems and a summary of existing methods. We discuss emerging topics, open issues, and provide our perspective on advancing the domain. The survey serves as introductory material for readers from academia and industry to the topic and identifies notable opportunities for further research.

Zero-Shot Learning (ZSL) aims to transfer classification capability from seen to unseen classes. Recent methods have proved that generalization and specialization are two essential abilities to achieve good performance in ZSL. However, focusing on only one of the abilities may result in models that are either too general with degraded classification ability or too specialized to generalize to unseen classes. In this paper, we propose an end-to-end network, termed as BGSNet, which equips and balances generalization and specialization abilities at the instance and dataset level. Specifically, BGSNet consists of two branches: the Generalization Network (GNet), which applies episodic meta-learning to learn generalized knowledge, and the Balanced Specialization Network (BSNet), which adopts multiple attentive extractors to extract discriminative features and achieve instance-level balance. A novel self-adjusted diversity loss is designed to optimize BSNet with redundancy reduced and diversity boosted. We further propose a differentiable dataset-level balance and update the weights in a linear annealing schedule to simulate network pruning and thus obtain the optimal structure for BSNet with dataset-level balance achieved. Experiments on four benchmark datasets demonstrate our model's effectiveness. Sufficient component ablations prove the necessity of integrating and balancing generalization and specialization abilities.

Language models (LMs) can reproduce (or amplify) toxic language seen during training, which poses a risk to their practical application. In this paper, we conduct extensive experiments to study this phenomenon. We analyze the impact of prompts, decoding strategies and training corpora on the output toxicity. Based on our findings, we propose a simple yet effective unsupervised method for language models to ``detoxify'' themselves without an additional large corpus or external discriminator. Compared to a supervised baseline, our proposed method shows better toxicity reduction with good generation quality in the generated content under multiple settings. Warning: some examples shown in the paper may contain uncensored offensive content.

In this paper, we propose SkipBERT to accelerate BERT inference by skipping the computation of shallow layers.To achieve this, our approach encodes small text chunks into independent representations, which are then materialized to approximate the shallow representation of BERT.Since the use of such approximation is inexpensive compared with transformer calculations, we leverage it to replace the shallow layers of BERT to skip their runtime overhead.With off-the-shelf early exit mechanisms, we also skip redundant computation from the highest few layers to further improve inference efficiency.Results on GLUE show that our approach can reduce latency by 65% without sacrificing performance.By using only two-layer transformer calculations, we can still maintain 95% accuracy of BERT. 1

High-quality phrase representations are essential to finding topics and related terms in documents (a.k.a.topic mining).Existing phrase representation learning methods either simply combine unigram representations in a contextfree manner or rely on extensive annotations to learn context-aware knowledge.In this paper, we propose UCTOPIC, a novel unsupervised contrastive learning framework for context-aware phrase representations and topic mining.UCTOPIC is pretrained in a large scale to distinguish if the contexts of two phrase mentions have the same semantics.The key to pretraining is positive pair construction from our phrase-oriented assumptions.However, we find traditional in-batch negatives cause performance decay when finetuning on a dataset with small topic numbers.Hence, we propose cluster-assisted contrastive learning (CCL) which largely reduces noisy negatives by selecting negatives from clusters and further improves phrase representations for topics accordingly.UCTOPIC outperforms the state-of-the-art phrase representation model by 38.2% NMI in average on four entity clustering tasks.Comprehensive evaluation on topic mining shows that UCTOPIC can extract coherent and diverse topical phrases.

Bodhisattwa Prasad Majumder, Harsh Jhamtani, Taylor Berg-Kirkpatrick, Julian McAuley. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

We present Knowledge Distillation with Meta Learning (MetaDistil), a simple yet effective alternative to traditional knowledge distillation (KD) methods where the teacher model is fixed during training.We show the teacher network can learn to better transfer knowledge to the student network (i.e., learning to teach) with the feedback from the performance of the distilled student network in a meta learning framework.Moreover, we introduce a pilot update mechanism to improve the alignment between the inner-learner and meta-learner in meta learning algorithms that focus on an improved inner-learner.Experiments on various benchmarks show that MetaDistil can yield significant improvements compared with traditional KD algorithms and is less sensitive to the choice of different student capacity and hyperparameters, facilitating the use of KD on different tasks and models. 1

Decision-making involves biases from past experiences, which are difficult to perceive and eliminate. We investigate a specific type of anchoring bias, in which decision-makers are anchored by their own recent decisions, e.g. a college admission officer sequentially reviewing students. We propose an algorithm that identifies existing anchored decisions, reduces sequential dependencies to previous decisions, and mitigates decision inaccuracies post-hoc with 2% increased agreement to ground-truth on a large-scale college admission decision data set. A crowd-sourced study validates this algorithm on product preferences (5% increased agreement). To avoid biased decisions ex-ante, we propose a procedure that presents instances in an order that reduces anchoring bias in real-time. Tested in another crowd-sourced study, it reduces bias and increases agreement to ground-truth by 7%. Our work reinforces individuals with similar characteristics to be treated similarly, independent of when they were reviewed in the decision-making process.

Sequential recommenders aim to capture users' dynamic interests from their historical action sequences, but remain challenging due to data sparsity issues, as well as the noisy and complex relationships among items in a sequence. Several approaches have sought to alleviate these issues using side-information, such as item content (e.g., images), action types (e.g., click, purchase). While useful, we argue one of the main contextual signals is largely ignored-namely users' queries. When users browse and consume products (e.g., music, movies), their sequential interactions are usually a combination of queries, clicks (etc.). Most interaction datasets discard queries, and corresponding methods simply model sequential behaviors over items and thus ignore this critical context of user interactions.

Knowledge-enhanced pre-trained models for language representation have been shown to be more effective in knowledge base construction tasks (i.e.,~relation extraction) than language models such as BERT. These knowledge-enhanced language models incorporate knowledge into pre-training to generate representations of entities or relationships. However, existing methods typically represent each entity with a separate embedding. As a result, these methods struggle to represent out-of-vocabulary entities and a large amount of parameters, on top of their underlying token models (i.e., the transformer), must be used and the number of entities that can be handled is limited in practice due to memory constraints. Moreover, existing models still struggle to represent entities and relationships simultaneously. To address these problems, we propose a new pre-trained model that learns representations of both entities and relationships from token spans and span pairs in the text respectively. By encoding spans efficiently with span modules, our model can represent both entities and their relationships but requires fewer parameters than existing models. We pre-trained our model with the knowledge graph extracted from Wikipedia and test it on a broad range of supervised and unsupervised information extraction tasks. Results show that our model learns better representations for both entities and relationships than baselines, while in supervised settings, fine-tuning our model outperforms RoBERTa consistently and achieves competitive results on information extraction tasks.

Prediction models can exhibit sensitivity with respect to training data: small changes in the training data can produce models that assign conflicting predictions to individual data points during test time. In this work, we study this sensitivity in recommender systems, where users' recommendations are drastically altered by minor perturbations in other unrelated users' interactions. We introduce a measure of stability for recommender systems, called Rank List Sensitivity (RLS), which measures how rank lists generated by a given recommender system at test time change as a result of a perturbation in the training data. We develop a method, CASPER, which uses cascading effect to identify the minimal and systematical perturbation to induce higher instability in a recommender system. Experiments on four datasets show that recommender models are overly sensitive to minor perturbations introduced randomly or via CASPER - even perturbing one random interaction of one user drastically changes the recommendation lists of all users.Importantly, with CASPER perturbation, the models generate more unstable recommend ations for low-accuracy users (i.e., those who receive low-quality recommendations) than high-accuracy ones.

The large population of home cooks with dietary restrictions is under-served by existing cooking resources and recipe generation models.To help them, we propose the task of controllable recipe editing: adapt a base recipe to satisfy a user-specified dietary constraint.This task is challenging, and cannot be adequately solved with human-written ingredient substitution rules or existing end-to-end recipe generation models.We tackle this problem with SHARE: a System for Hierarchical Assistive Recipe Editing, which performs simultaneous ingredient substitution before generating natural-language steps using the edited ingredients.By decoupling ingredient and step editing, our step generator can explicitly integrate the available ingredients.Experiments on the novel RecipePairs dataset-83K pairs of similar recipes where each recipe satisfies one of seven dietary constraints-demonstrate that SHARE produces convincing, coherent recipes that are appropriate for a target dietary constraint.We further show through human evaluations and real-world cooking trials that recipes edited by SHARE can be easily followed by home cooks to create appealing dishes.

Masked language modeling is widely used for pretraining large language models for natural language understanding (NLU).However, random masking is suboptimal, allocating an equal masking rate for all tokens.In this paper, we propose InforMask, a new unsupervised masking strategy for training masked language models.InforMask exploits Pointwise Mutual Information (PMI) to select the most informative tokens to mask.We further propose two optimizations for InforMask to improve its efficiency.With a one-off preprocessing step, InforMask outperforms random masking and previously proposed masking strategies on the factual recall benchmark LAMA and the question answering benchmark SQuAD v1 and v2. 1

Intermediate-task transfer can benefit a wide range of NLP tasks with properly selected source datasets.However, it is computationally infeasible to experiment with all intermediate transfer combinations, making choosing a useful source task a challenging problem.In this paper, we anticipate that task-specific parameters updated in parameter-efficient tuning methods are likely to encode task-specific information.Therefore, such parameters can be predictive for inter-task transferability.Thus, we propose to exploit these efficiently tuned parameters as off-the-shelf task embeddings for the efficient selection of source datasets for intermediate-task transfer.We experiment with 11 text classification tasks and 11 question answering tasks.Experimental results show that our approach can consistently outperform existing inter-task transferability prediction methods while being conceptually simple and computationally efficient.Our analysis also reveals that the ability of efficiently tuned parameters on transferability prediction is disentangled with their in-task performance.This allows us to use parameters from early checkpoints as task embeddings to further improve efficiency. 1

Music performance synthesis aims to synthesize a musical score into a natural performance. In this paper, we borrow recent advances in text-to-speech synthesis and present the Deep Performer—a novel system for score-to-audio music performance synthesis. Unlike speech, music often contains polyphony and long notes. Hence, we propose two new techniques for handling polyphonic inputs and providing a fine-grained conditioning in a transformer encoder-decoder model. To train our proposed system, we present a new violin dataset consisting of paired recordings and scores along with estimated alignments between them. We show that our proposed model can synthesize music with clear polyphony and harmonic structures. In a listening test, we achieve competitive quality against the baseline model, a conditional generative audio model, in terms of pitch accuracy, timbre and noise level. Moreover, our proposed model significantly outperforms the baseline on an existing piano dataset in overall quality.

Models that generate extractive rationales (i.e., subsets of features) or natural language explanations (NLEs) for their predictions are important for explainable AI. While an extractive rationale provides a quick view of the features most responsible for a prediction, an NLE allows for a comprehensive description of the decision-making process behind a prediction. However, current models that generate the best extractive rationales or NLEs often fall behind the state-of-the-art (SOTA) in terms of task performance. In this work, we bridge this gap by introducing RExC, a self-rationalizing framework that grounds its predictions and two complementary types of explanations (NLEs and extractive rationales) in background knowledge. Our framework improves over previous methods by: (i) reaching SOTA task performance while also providing explanations, (ii) providing two types of explanations, while existing models usually provide only one type, and (iii) beating by a large margin the previous SOTA in terms of quality of both types of explanations. Furthermore, a perturbation analysis in RExC shows a high degree of association between explanations and predictions, a necessary property of faithful explanations.

Recommender systems (RecSys) play important roles in helping users navigate, discover, and consume large and highly dynamic information. Today, many RecSys solutions deployed in the real world rely on categorical user-profiles and/or pre-calculated recommendation actions that stay static during a user session. However, recent trends suggest that RecSys need to model user intent in real time and constantly adapt to meet user needs at the moment or change user behavior in situ. There are three primary drivers for this emerging need of online adaptation. First, in order to meet the increasing demand for a better personalized experience, the personalization dimensions and space will grow larger and larger. It would not be feasible to pre-compute recommended actions for all personalization scenarios beyond a certain scale. Second, in many settings the system does not have user prior history to leverage. Estimating user intent in real time is the only feasible way to personalize. As various consumer privacy laws tighten, it is foreseeable that many businesses will reduce their reliance on static user profiles. Therefore, it makes the modeling of user intent in real time an important research topic. Third, a user's intent often changes within a session and between sessions, and user behavior could shift significantly during dramatic events. A RecSys should adapt in real time to meet user needs and be robust against distribution shifts. The online and adaptive recommender systems (OARS) workshop offers a focused discussion of the study and application of OARS, and will bring together an interdisciplinary community of researchers and practitioners from both industry and academia. KDD, as the premier data science conference, is an ideal venue to gather leaders in the field to further research into OARS and promote its adoption. This workshop is complementary to several sessions of the main conference (e.g., recommendation, reinforcement learning, etc.) and brings them together using a practical and focused application.

Prompt-based learning (i.e., prompting) is an emerging paradigm for exploiting knowledge learned by a pretrained language model.In this paper, we propose Automatic Multi-Label Prompting (AMuLaP), a simple yet effective method to automatically select label mappings for few-shot text classification with prompting.Our method exploits one-to-many label mappings and a statistics-based algorithm to select label mappings given a prompt template.Our experiments demonstrate that AMu-LaP achieves competitive performance on the GLUE benchmark without human effort or external resources.1

We leverage the Neural Tangent Kernel and its equivalence to training infinitely-wide neural networks to devise $\infty$-AE: an autoencoder with infinitely-wide bottleneck layers. The outcome is a highly expressive yet simplistic recommendation model with a single hyper-parameter and a closed-form solution. Leveraging $\infty$-AE's simplicity, we also develop Distill-CF for synthesizing tiny, high-fidelity data summaries which distill the most important knowledge from the extremely large and sparse user-item interaction matrix for efficient and accurate subsequent data-usage like model training, inference, architecture search, etc. This takes a data-centric approach to recommendation, where we aim to improve the quality of logged user-feedback data for subsequent modeling, independent of the learning algorithm. We particularly utilize the concept of differentiable Gumbel-sampling to handle the inherent data heterogeneity, sparsity, and semi-structuredness, while being scalable to datasets with hundreds of millions of user-item interactions. Both of our proposed approaches significantly outperform their respective state-of-the-art and when used together, we observe $96-105$% of $\infty$-AE's performance on the full dataset with as little as $0.1$% of the original dataset size, leading us to explore the counter-intuitive question: Is more data what you need for better recommendation?

Bundle recommender systems recommend sets of items (e.g., pants, shirt, and shoes) to users, but they often suffer from two issues: significant interaction sparsity and a large output space. In this work, we extend multi-round conversational recommendation (MCR) to alleviate these issues. MCR—which uses a conversational paradigm to elicit user interests by asking user preferences on tags (e.g., categories or attributes) and handling user feedback across multiple rounds—is an emerging recommendation setting to acquire user feedback and narrow down the output space, but has not been explored in the context of bundle recommendation.

Conversational recommender systems offer a way for users to engage in multi-turn conversations to find items they enjoy. For users to trust an agent and give effective feedback, the recommender system must be able to explain its suggestions and rationales. We develop a two-part framework for training multi-turn conversational recommenders that provide recommendation rationales that users can effectively interact with to receive better recommendations. First, we train a recommender system to jointly suggest items and explain its reasoning via subjective rationales. We then fine-tune this model to incorporate iterative user feedback via self-supervised bot-play. Experiments on three real-world datasets demonstrate that our system can be applied to different recommendation models across diverse domains to achieve state-of-the-art performance in multi-turn recommendation. Human studies show that systems trained with our framework provide more useful, helpful, and knowledgeable suggestions in warm- and cold-start settings. Our framework allows us to use only product reviews during training, avoiding the need for expensive dialog transcript datasets that limit the applicability of previous conversational recommender agents.

Online recommendation requires handling rapidly changing user preferences. Deep reinforcement learning (DRL) is an effective means of capturing users' dynamic interest during interactions with recommender systems. Generally, it is challenging to train a DRL agent in online recommender systems because of the sparse rewards caused by the large action space (e.g., candidate item space) and comparatively fewer user interactions. Leveraging experience replay (ER) has been extensively studied to conquer the issue of sparse rewards. However, they adapt poorly to the complex environment of online recommender systems and are inefficient in learning an optimal strategy from past experience. As a step to filling this gap, we propose a novel state-aware experience replay model, in which the agent selectively discovers the most relevant and salient experiences and is guided to find the optimal policy for online recommendations. In particular, a locality-sensitive hashing method is proposed to selectively retain the most meaningful experience at scale and a prioritized reward-driven strategy is designed to replay more valuable experiences with higher chance. We formally show that the proposed method guarantees the upper and lower bound on experience replay and optimizes the space complexity, as well as empirically demonstrate our model's superiority to several existing experience replay methods over three benchmark simulation platforms.

Sequential recommendation aims to model dynamic user behavior from historical interactions. Self-attentive methods have proven effective at capturing short-term dynamics and long-term preferences. Despite their success, these approaches still struggle to model sparse data, on which they struggle to learn high-quality item representations. We propose to model user dynamics from shopping intents and interacted items simultaneously. The learned intents are coarse-grained and work as prior knowledge for item recommendation. To this end, we present a coarse-to-fine self-attention framework, namely CaFe, which explicitly learns coarse-grained and fine-grained sequential dynamics. Specifically, CaFe first learns intents from coarse-grained sequences which are dense and hence provide high-quality user intent representations. Then, CaFe fuses intent representations into item encoder outputs to obtain improved item representations. Finally, we infer recommended items based on representations of items and corresponding intents. Experiments on sparse datasets show that CaFe outperforms state-of-the-art self-attentive recommenders by 44.03% [email protected] on average.

With the abundance of large-scale deep learning models, it has become possible to repurpose pre-trained networks for new tasks. Recent works on adversarial reprogramming have shown that it is possible to repurpose neural networks for alternate tasks without modifying the network architecture or parameters. However these works only consider original and target tasks within the same data domain. In this work, we broaden the scope of adversarial reprogramming beyond the data modality of the original task. We analyze the feasibility of adversarially repurposing image classification neural networks for Natural Language Processing (NLP) and other sequence classification tasks. We design an efficient adversarial program that maps a sequence of discrete tokens into an image which can be classified to the desired class by an image classification model. We demonstrate that by using highly efficient adversarial programs, we can reprogram image classifiers to achieve competitive performance on a variety of text and sequence classification benchmarks without retraining the network.

We study the practical consequences of dataset sampling strategies on the ranking performance of recommendation algorithms. Recommender systems are generally trained and evaluated on samples of larger datasets. Samples are often taken in a naive or ad-hoc fashion: e.g. by sampling a dataset randomly or by selecting users or items with many interactions. As we demonstrate, commonly-used data sampling schemes can have significant consequences on algorithm performance. Following this observation, this paper makes three main contributions: (1) characterizing the effect of sampling on algorithm performance, in terms of algorithm and dataset characteristics (e.g. sparsity characteristics, sequential dynamics, etc.); (2) designing SVP-CF, which is a data-specific sampling strategy, that aims to preserve the relative performance of models after sampling, and is especially suited to long-tailed interaction data; and (3) developing an oracle, DATA-GENIE, which can suggest the sampling scheme that is most likely to preserve model performance for a given dataset. The main benefit of DATA-GENIE is that it will allow recommender system practitioners to quickly prototype and compare various approaches, while remaining confident that algorithm performance will be preserved, once the algorithm is retrained and deployed on the complete data. Detailed experiments show that using DATA-GENIE, we can discard upto 5x more data than any sampling strategy with the same level of performance.

Users’ interactions with items are driven by various intents (e.g., preparing for holiday gifts, shopping for fishing equipment, etc.). However, users’ underlying intents are often unobserved/latent, making it challenging to leverage such latent intents for Sequential recommendation (SR). To investigate the benefits of latent intents and leverage them effectively for recommendation, we propose Intent Contrastive Learning (ICL), a general learning paradigm that leverages a latent intent variable into SR. The core idea is to learn users’ intent distribution functions from unlabeled user behavior sequences and optimize SR models with contrastive self-supervised learning (SSL) by considering the learnt intents to improve recommendation. Specifically, we introduce a latent variable to represent users’ intents and learn the distribution function of the latent variable via clustering. We propose to leverage the learnt intents into SR models via contrastive SSL, which maximizes the agreement between a view of sequence and its corresponding intent. The training is alternated between intent representation learning and the SR model optimization steps within the generalized expectation-maximization (EM) framework. Fusing user intent information into SR also improves model robustness. Experiments conducted on four real-world datasets demonstrate the superiority of the proposed learning paradigm, which improves performance, and robustness against data sparsity and noisy interaction issues 1.

Zero-Shot Learning (ZSL) aims to transfer learned knowledge from observed classes to unseen classes via semantic correlations. A promising strategy is to learn a global-local representation that incorporates global information with extra localities (i.e., small parts/regions of inputs). However, existing methods discover localities based on explicit features without digging into the inherent properties and relationships among regions. In this work, we propose a novel Entropy-guided Reinforced Partial Convolutional Network (ERPCNet), which extracts and aggregates localities progressively based on semantic relevance and visual correlations without human-annotated regions. ERPCNet uses reinforced partial convolution and entropy guidance; it not only discovers global-cooperative localities dynamically but also converges faster for policy gradient optimization. We conduct extensive experiments to demonstrate ERPCNet’s performance through comparisons with state-of-the-art methods under ZSL and Generalized Zero-Shot Learning (GZSL) settings on four benchmark datasets. We also show that ERPCNet is time efficient and explainable through visualization analysis.

Jiacheng Li, Haibo Ding, Jingbo Shang, Julian McAuley, Zhe Feng. 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.

Voice cloning is the task of learning to synthesize the voice of an unseen speaker from a few samples. While current voice cloning methods achieve promising results in Text-to-Speech (TTS) synthesis for a new voice, these approaches lack the ability to control the expressiveness of synthesized audio. In this work, we propose a controllable voice cloning method that allows fine-grained control over various style aspects of the synthesized speech for an unseen speaker. We achieve this by explicitly conditioning the speech synthesis model on a speaker encoding, pitch contour and latent style tokens during training. Through both quantitative and qualitative evaluations, we show that our framework can be used for various expressive voice cloning tasks using only a few transcribed or untranscribed speech samples for a new speaker. These cloning tasks include style transfer from a reference speech, synthesizing speech directly from text, and fine-grained style control by manipulating the style conditioning variables during inference.

Recently, self-attentive models have shown promise in sequential recommendation, given their potential to capture user long-term preferences and short-term dynamics simultaneously. Despite their success, we argue that self-attention modules, as a non-local operator, often fail to capture short-term user dynamics accurately due to a lack of inductive local bias. To examine our hypothesis, we conduct an analytical experiment on controlled 'short-term' scenarios. We observe a significant performance gap between self-attentive recommenders with and without local constraints, which implies that short-term user dynamics are not sufficiently learned by existing self-attentive recommenders. Motivated by this observation, we propose a simple framework, (Locker) for self-attentive recommenders in a plug-and-play fashion. By combining the proposed local encoders with existing global attention heads, Locker enhances short-term user dynamics modeling, while retaining the long-term semantics captured by standard self-attentive encoders. We investigate Locker with five different local methods, outperforming state-of-the-art self-attentive recom- menders on three datasets by 17.19% ([email protected]) on average.

Shuyang Li, Jin Cao, Mukund Sridhar, Henghui Zhu, Shang-Wen Li, Wael Hamza, Julian McAuley. Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume. 2021.

Recent studies on compression of pretrained language models (e.g., BERT) usually use preserved accuracy as the metric for evaluation.In this paper, we propose two new metrics, label loyalty and probability loyalty that measure how closely a compressed model (i.e., student) mimics the original model (i.e., teacher).We also explore the effect of compression with regard to robustness under adversarial attacks.We benchmark quantization, pruning, knowledge distillation and progressive module replacing with loyalty and robustness.By combining multiple compression techniques, we provide a practical strategy to achieve better accuracy, loyalty and robustness. 1

Many recommender systems deployed in the real world rely on categorical user-profiles and/or pre-calculated recommendation actions that stay static during a user session. Recent trends suggest that recommender systems should model user intent in real time and constantly adapt to meet user needs at the moment or change user behavior in situ. In addition, there have been many advances that make online and adaptive recommender systems (OARS) feasible, scalable, and more sophisticated. This workshop aims to bring together practitioners and researchers from academia and industry to discuss the challenges and approaches to implement OARS algorithms and systems and improve user experiences by better modeling and responding to user intent.

Bodhisattwa Prasad Majumder, Sudha Rao, Michel Galley, Julian McAuley. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Canwen Xu, Wangchunshu Zhou, Tao Ge, Ke Xu, Julian McAuley, Furu Wei. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Live-streaming platforms broadcast user-generated video in real-time. Recommendation on these platforms shares similarities with traditional settings, such as a large volume of heterogeneous content and highly skewed interaction distributions. However, several challenges must be overcome to adapt recommendation algorithms to live-streaming platforms: first, content availability is dynamic which restricts users to choose from only a subset of items at any given time; during training and inference we must carefully handle this factor in order to properly account for such signals, where ‘non-interactions’ reflect availability as much as implicit preference. Streamers are also fundamentally different from ‘items’ in traditional settings: repeat consumption of specific channels plays a significant role, though the content itself is fundamentally ephemeral.

We investigate whether model extraction can be used to ‘steal’ the weights of sequential recommender systems, and the potential threats posed to victims of such attacks. This type of risk has attracted attention in image and text classification, but to our knowledge not in recommender systems. We argue that sequential recommender systems are subject to unique vulnerabilities due to the specific autoregressive regimes used to train them. Unlike many existing recommender attackers, which assume the dataset used to train the victim model is exposed to attackers, we consider a data-free setting, where training data are not accessible. Under this setting, we propose an API-based model extraction method via limited-budget synthetic data generation and knowledge distillation. We investigate state-of-the-art models for sequential recommendation and show their vulnerability under model extraction and downstream attacks.

Deep networks often suffer from vanishing or exploding gradients due to inefficient signal propagation, leading to long training times or convergence difficulties. Various architecture designs, sophisticated residual-style networks, and initialization schemes have been shown to improve deep signal propagation. Recently, Pennington et al. [2017] used free probability theory to show that dynamical isometry plays an integral role in efficient deep learning. We show that the simplest architecture change of gating each residual connection using a single zero-initialized parameter satisfies initial dynamical isometry and outperforms more complex approaches. Although much simpler than its predecessors, this gate enables training thousands of fully connected layers with fast convergence and better test performance for ResNets trained on an image recognition task. We apply this technique to language modeling and find that we can easily train 120-layer Transformers. When applied to 12 layer Transformers, it converges 56% faster.

Recent advances in video manipulation techniques have made the generation of fake videos more accessible than ever before. Manipulated videos can fuel disinformation and reduce trust in media. Therefore detection of fake videos has garnered immense interest in academia and industry. Recently developed Deepfake detection methods rely on Deep Neural Networks (DNNs) to distinguish AI-generated fake videos from real videos. In this work, we demonstrate that it is possible to bypass such detectors by adversarially modifying fake videos synthesized using existing Deepfake generation methods. We further demonstrate that our adversarial perturbations are robust to image and video compression codecs, making them a real-world threat. We present pipelines in both white-box and black-box attack scenarios that can fool DNN based Deepfake detectors into classifying fake videos as real.

Existing persona-grounded dialog models often fail to capture simple implications of given persona descriptions, something which humans are able to do seamlessly.For example, state-of-the-art models cannot infer that interest in hiking might imply love for nature or longing for a break.In this paper, we propose to expand available persona sentences using existing commonsense knowledge bases and paraphrasing resources to imbue dialog models with access to an expanded and richer set of persona descriptions.Additionally, we introduce fine-grained grounding on personas by encouraging the model to make a discrete choice among persona sentences while synthesizing a dialog response.Since such a choice is not observed in the data, we model it using a discrete latent random variable and use variational learning to sample from hundreds of persona expansions.Our model outperforms competitive baselines on the PERSONA-CHAT dataset in terms of dialog quality and diversity while achieving persona-consistent and controllable dialog generation.

In this work, we perform the first large-scale analysis of discourse in media dialog and its impact on generative modeling of dialog turns, with a focus on interrogative patterns and use of external knowledge.Discourse analysis can help us understand modes of persuasion, entertainment, and information elicitation in such settings, but has been limited to manual review of small corpora.We introduce Interview-a large-scale (105K conversations) media dialog dataset collected from news interview transcripts-which allows us to investigate such patterns at scale.We present a dialog model that leverages external knowledge as well as dialog acts via auxiliary losses and demonstrate that our model quantitatively and qualitatively outperforms strong discourse-agnostic baselines for dialog modeling-generating more specific and topical responses in interview-style conversations.

Over a decade ago, the CS1 course for students without prior programming experience at a large research-intensive university was redesigned to incorporate three best practices in teaching programming: Media Computation, Pair Programming, and Peer Instruction. The purpose of this revision was to improve the quality of the course, appeal to a larger student body, and improve retention in the major. An initial analysis of the course indicated an increase in pass rates and 1-yr retention of students in the major. Now that time has passed and those students impacted by the revision have had time to graduate, this longitudinal study revisits and expands on these prior findings through examining student outcomes over a twelve year period (2001 through 2013). The student outcomes examined include failure rates in CS1, retention rates in the major, rates of switching into the major, time to degree, and performance in subsequent major courses. We compare these findings against similar metrics collected for another CS1 course at the same institution that caters to students with prior programming experience and did not make changes during this same time period. Overall, the inclusion of media computation, pair programming, and peer instruction corresponds to a significant improvement in passing rates for CS1 as well as retention of majors from CS1 through graduation. In turn, there is no indication that this larger group of students experienced any harm in terms of lower grades in upper-division courses or their time to degree.

Speech recognition (ASR) and speaker diarization (SD) models have traditionally been trained separately to produce rich conversation transcripts with speaker labels. Recent advances have shown that joint ASR and SD models can learn to leverage audio-lexical inter-dependencies to improve word diarization performance. We introduce a new benchmark of hour-long podcasts collected from the weekly This American Life radio program to better compare these approaches when applied to extended multi-speaker conversations. We find that training separate ASR and SD models perform better when utterance boundaries are known but otherwise joint models can perform better. To handle long conversations with unknown utterance boundaries, we introduce a striding attention decoding algorithm and data augmentation techniques which, combined with model pre-training, improves ASR and SD.

In this paper, we propose Patience-based Early Exit, a straightforward yet effective inference method that can be used as a plug-and-play technique to simultaneously improve the efficiency and robustness of a pretrained language model (PLM). To achieve this, our approach couples an internal-classifier with each layer of a PLM and dynamically stops inference when the intermediate predictions of the internal classifiers do not change for a pre-defined number of steps. Our approach improves inference efficiency as it allows the model to make a prediction with fewer layers. Meanwhile, experimental results with an ALBERT model show that our method can improve the accuracy and robustness of the model by preventing it from overthinking and exploiting multiple classifiers for prediction, yielding a better accuracy-speed trade-off compared to existing early exit methods.

Graph-based recommendation algorithms treat user-item interactions as bipartite graphs, based on which low-dimensional vector representations of users and items seek to preserve the relationships among them. Previous methods usually capture users' preferences by directly learning first-order neighborhood patterns for each node, which limits their ability to exploit the similarity between two distant users/items as well as a user's preferences toward distant items. To address this potential weakness, in this paper, we propose SMOG-CF (Stacked Mixed-Order Graph Convolutional Networks for Collaborative Filtering), a GCN-based framework that can directly capture high-order connectivity among nodes. Instead of implicitly capturing high-order connectivity through embedding propagation, SMOG-CF facilitates ‘path-level’ information propagation between neighboring nodes at any order. The matrix form of our embedding propagation formulas yields a model that is easy to deploy and can be extended to a general framework by adopting various information construction and aggregation equations. Experiments on several datasets of varying scale demonstrate the efficacy of our model.

We investigate a growing body of work that seeks to improve recommender systems through the use of review text. Generally, these papers argue that since reviews 'explain' users' opinions, they ought to be useful to infer the underlying dimensions that predict ratings or purchases. Schemes to incorporate reviews range from simple regularizers to neural network approaches. Our initial findings reveal several discrepancies in reported results, partly due to (e.g.) copying results across papers despite changes in experimental settings or data pre-processing. First, we attempt a comprehensive analysis to resolve these ambiguities. Further investigation calls for discussion on a much larger problem about the "importance" of user reviews for recommendation. Through a wide range of experiments, we observe several cases where state-of-the-art methods fail to outperform existing baselines, especially as we deviate from a few narrowly-defined settings where reviews are useful. We conclude by providing hypotheses for our observations, that seek to characterize under what conditions reviews are likely to be helpful. Through this work, we aim to evaluate the direction in which the field is progressing and encourage robust empirical evaluation.

Sequential recommender systems seek to exploit the order of users' interactions, in order to predict their next action based on the context of what they have done recently. Traditionally, Markov Chains(MCs), and more recently Recurrent Neural Networks (RNNs) and Self Attention (SA) have proliferated due to their ability to capture the dynamics of sequential patterns. However a simplifying assumption made by most of these models is to regard interaction histories as ordered sequences, without regard for the time intervals between each interaction (i.e., they model the time-order but not the actual timestamp). In this paper, we seek to explicitly model the timestamps of interactions within a sequential modeling framework to explore the influence of different time intervals on next item prediction. We propose TiSASRec (Time Interval aware Self-attention based sequential recommendation), which models both the absolute positions of items as well as the time intervals between them in a sequence. Extensive empirical studies show the features of TiSASRec under different settings and compare the performance of self-attention with different positional encodings. Furthermore, experimental results show that our method outperforms various state-of-the-art sequential models on both sparse and dense datasets and different evaluation metrics.

Modern collaborative filtering algorithms seek to provide personalized product recommendations by uncovering patterns in consumer-product interactions. However, these interactions can be biased by how the product is marketed, for example due to the selection of a particular human model in a product image. These correlations may result in the underrepresentation of particular niche markets in the interaction data; for example, a female user who would potentially like motorcycle products may be less likely to interact with them if they are promoted using stereotypically 'male' images.

Users exhibit different intents across e-commerce services (e.g. discovering items, purchasing gifts, etc.) which drives them to interact with a wide variety of items in multiple ways (e.g. click, add-to-cart, add-to-favorites, purchase). To give better recommendations, it is important to capture user intent, in addition to considering their historic interactions. However these intents are by definition latent, as we observe only a user’s interactions, and not their underlying intent. To discover such latent intents, and use them effectively for recommendation, in this paper we propose an Attentive Sequential model of Latent Intent (ASLI in short). Our model first learns item similarities from users’ interaction histories via a self-attention layer, then uses a Temporal Convolutional Network layer to obtain a latent representation of the user’s intent from her actions on a particular category. We use this representation to guide an attentive model to predict the next item. Results from our experiments show that our model can capture the dynamics of user behavior and preferences, leading to state-of-the-art performance across datasets from two major e-commerce platforms, namely Etsy and Alibaba.

This paper presents computational approaches for automatically detecting critical plot twists in reviews of media products.First, we created a large-scale book review dataset that includes fine-grained spoiler annotations at the sentence-level, as well as book and (anonymized) user information.Second, we carefully analyzed this dataset, and found that: spoiler language tends to be book-specific; spoiler distributions vary greatly across books and review authors; and spoiler sentences tend to jointly appear in the latter part of reviews.Third, inspired by these findings, we developed an end-to-end neural network architecture to detect spoiler sentences in review corpora.Quantitative and qualitative results demonstrate that the proposed method substantially outperforms existing baselines.• This was a perfect, albeit bloody, end to the series.• Though there were deaths that were definitely unwarranted: Fred Hedwig Moody Tonks Lupin Dobby, there were some really heartfelt and memorable moments: Narcissa saving Harry, Ron coming back, Hermione and Ron, Harry and Ginny, Molly killing Bellatrix, etc. • I wish we could have spent more time at Hogwarts, as one of my favorite characters, the amazing Minerva McGonagall, resides there, and we couldn't see more of her amazingness in the Battle of Hogwarts.• Harry Potter was a really, really great series that I think will be (and is) timeless.

Generating the Top-N recommendations from a large corpus is computationally expensive to perform at scale. Candidate generation and re-ranking based approaches are often adopted in industrial settings to alleviate efficiency problems. However it remains to be fully studied how well such schemes approximate complete rankings (or how many candidates are required to achieve a good approximation), or to develop systematic approaches to generate high-quality candidates efficiently. In this paper, we seek to investigate these questions via proposing a candidate generation and re-ranking based framework (CIGAR), which first learns a preference-preserving binary embedding for building a hash table to retrieve candidates, and then learns to re-rank the candidates using real-valued ranking models with a candidate-oriented objective. We perform a comprehensive study on several large-scale real-world datasets consisting of millions of users/items and hundreds of millions of interactions. Our results show that CIGAR significantly boosts the Top-N accuracy against state-of-the-art recommendation models, while reducing the query time by orders of magnitude. We hope that this work could draw more attention to the candidate generation problem in recommender systems.

Sequential patterns play an important role in building modern recommender systems. To this end, several recommender systems have been built on top of Markov Chains and Recurrent Models (among others). Although these sequential models have proven successful at a range of tasks, they still struggle to uncover complex relationships nested in user purchase histories. In this paper, we argue that modeling pairwise relationships directly leads to an efficient representation of sequential features and captures complex item correlations. Specifically, we propose a 2D convolutional network for sequential recommendation (CosRec). It encodes a sequence of items into a three-way tensor; learns local features using 2D convolutional filters; and aggregates high-order interactions in a feedforward manner. Quantitative results on two public datasets show that our method outperforms both conventional methods and recent sequence-based approaches, achieving state-of-the-art performance on various evaluation metrics.

Modeling fashion compatibility is challenging due to its complexity and subjectivity. Existing work focuses on predicting compatibility between product images (e.g. an image containing a t-shirt and an image containing a pair of jeans). However, these approaches ignore real-world `scene' images (e.g. selfies); such images are hard to deal with due to their complexity, clutter, variations in lighting and pose (etc.) but on the other hand could potentially provide key context (e.g. the user's body type, or the season) for making more accurate recommendations. In this work, we propose a new task called `Complete the Look', which seeks to recommend visually compatible products based on scene images. We design an approach to extract training data for this task, and propose a novel way to learn the scene-product compatibility from fashion or interior design images. Our approach measures compatibility both globally and locally via CNNs and attention mechanisms. Extensive experiments show that our method achieves significant performance gains over alternative systems. Human evaluation and qualitative analysis are also conducted to further understand model behavior. We hope this work could lead to useful applications which link large corpora of real-world scenes with shoppable products.

Bodhisattwa Prasad Majumder, Shuyang Li, Jianmo Ni, Julian McAuley. 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.

Huanru Henry Mao, Bodhisattwa Prasad Majumder, Julian McAuley, Garrison Cottrell. 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.

Jianmo Ni, Jiacheng Li, Julian McAuley. 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.

Liliang Ren, Jianmo Ni, Julian McAuley. 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.

Recent approaches in text-to-speech (TTS) synthesis employ neural network strategies to vocode perceptually-informed spectrogram representations directly into listenable waveforms. Such vocoding procedures create a computational bottleneck in modern TTS pipelines. We propose an alternative approach which utilizes generative adversarial networks (GANs) to learn mappings from perceptually-informed spectrograms to simple magnitude spectrograms which can be heuristically vocoded. Through a user study, we show that our approach significantly outperforms naïve vocoding strategies while being hundreds of times faster than neural network vocoders used in state-of-the-art TTS systems. We also show that our method can be used to achieve state-of-the-art results in unsupervised synthesis of individual words of speech.

In this work, we demonstrate the existence of universal adversarial audio perturbations that cause mis-transcription of audio signals by automatic speech recognition (ASR) systems. We propose an algorithm to find a single quasi-imperceptible perturbation, which when added to any arbitrary speech signal, will most likely fool the victim speech recognition model. Our experiments demonstrate the application of our proposed technique by crafting audio-agnostic universal perturbations for the state-of-the-art ASR system -- Mozilla DeepSpeech. Additionally, we show that such perturbations generalize to a significant extent across models that are not available during training, by performing a transferability test on a WaveNet based ASR system.

Activity logs collected from wearable devices (e.g. Apple Watch, Fitbit, etc.) are a promising source of data to facilitate a wide range of applications such as personalized exercise scheduling, workout recommendation, and heart rate anomaly detection. However, such data are heterogeneous, noisy, diverse in scale and resolution, and have complex interdependencies, making them challenging to model. In this paper, we develop context-aware sequential models to capture the personalized and temporal patterns of fitness data. Specifically, we propose FitRec - an LSTM-based model that captures two levels of context information: context within a specific activity, and context across a user's activity history. We are specifically interested in (a) estimating a user's heart rate profile for a candidate activity; and (b) predicting and recommending suitable activities on this basis. We evaluate our model on a novel dataset containing over 250 thousand workout records coupled with hundreds of millions of parallel sensor measurements (e.g. heart rate, GPS) and metadata. We demonstrate that the model is able to learn contextual, personalized, and activity-specific dynamics of users' heart rate profiles during exercise. We evaluate the proposed model against baselines on several personalized recommendation tasks, showing the promise of using wearable data for activity modeling and recommendation.

Recommender Systems have proliferated as general-purpose approaches to model a wide variety of consumer interaction data. Specific instances make use of signals ranging from user feedback, item relationships, geographic locality, social influence (etc.). Typically, research proceeds by showing that making use of a specific signal (within a carefully designed model) allows for higher-fidelity recommendations on a particular dataset. Of course, the real situation is more nuanced, in which a combination of many signals may be at play, or favored in different proportion by individual users. Here we seek to develop a framework that is capable of combining such heterogeneous item relationships by simultaneously modeling (a) what modality of recommendation is a user likely to be susceptible to at a particular point in time; and (b) what is the best recommendation from each modality. Our method borrows ideas from mixtures-of-experts approaches as well as knowledge graph embeddings. We find that our approach naturally yields more accurate recommendations than alternatives, while also providing intuitive 'explanations' behind the recommendations it provides.

We study the problem of representing and recommending products for grocery shopping. We carefully investigate grocery transaction data and observe three important patterns: products within the same basket complement each other in terms of functionality (complementarity); users tend to purchase products that match their preferences (compatibility); and a significant fraction of users repeatedly purchase the same products over time (loyalty). Unlike conventional e-commerce settings, complementarity and loyalty are particularly predominant in the grocery shopping domain. This motivates a new representation learning approach to leverage complementarity and compatibility holistically, as well as a new recommendation approach to explicitly account for users' 'must-buy' purchases in addition to their overall preferences and needs. Doing so not only improves product classification and recommendation performance on both public and proprietary transaction data covering various grocery store types, but also reveals interesting findings about the relationships between preferences, necessity, and loyalty in consumer purchases.

Sequential dynamics are a key feature of many modern recommender systems, which seek to capture the 'context' of users' activities on the basis of actions they have performed recently. To capture such patterns, two approaches have proliferated: Markov Chains (MCs) and Recurrent Neural Networks (RNNs). Markov Chains assume that a user's next action can be predicted on the basis of just their last (or last few) actions, while RNNs in principle allow for longer-term semantics to be uncovered. Generally speaking, MC-based methods perform best in extremely sparse datasets, where model parsimony is critical, while RNNs perform better in denser datasets where higher model complexity is affordable. The goal of our work is to balance these two goals, by proposing a self-attention based sequential model (SASRec) that allows us to capture long-term semantics (like an RNN), but, using an attention mechanism, makes its predictions based on relatively few actions (like an MC). At each time step, SASRec seeks to identify which items are 'relevant' from a user's action history, and use them to predict the next item. Extensive empirical studies show that our method outperforms various state-of-the-art sequential models (including MC/CNN/RNN-based approaches) on both sparse and dense datasets. Moreover, the model is an order of magnitude more efficient than comparable CNN/RNN-based models. Visualizations on attention weights also show how our model adaptively handles datasets with various density, and uncovers meaningful patterns in activity sequences.

Modeling the complex interactions between users and items is at the core of designing successful recommender systems. One key task consists of predicting users’ personalized sequential behavior, where the challenge mainly lies in modeling ‘third-order’ interactions between a user, her previously visited item(s), and the next item to consume. In this paper, we propose a unified method, TransRec, to model such interactions for largescale sequential prediction. Methodologically, we embed items into a ‘transition space’ where users are modeled as translation vectors operating on item sequences. Empirically, this approach outperforms the state-of-the-art on a wide spectrum of real-world datasets.

Following precedent in employment discrimination law, two notions of disparity are widely-discussed in papers on fairness and ML. Algorithms exhibit treatment disparity if they formally treat members of protected subgroups differently; algorithms exhibit impact disparity when outcomes differ across subgroups (even unintentionally). Naturally, we can achieve impact parity through purposeful treatment disparity. One line of papers aims to reconcile the two parities proposing disparate learning processes (DLPs). Here, the sensitive feature is used during training but a group-blind classifier is produced. In this paper, we show that: (i) when sensitive and (nominally) nonsensitive features are correlated, DLPs will indirectly implement treatment disparity, undermining the policy desiderata they are designed to address; (ii) when group membership is partly revealed by other features, DLPs induce within-class discrimination; and (iii) in general, DLPs provide suboptimal trade-offs between accuracy and impact parity. Experimental results on several real-world datasets highlight the practical consequences of applying DLPs.

Building price- and budget-aware recommender systems is critical in settings where one wishes to produce recommendations that balance users' preferences (what they like) with a model of purchase likelihood (what they will buy). A trivial solution consists of learning global budget terms for each user based on their past expenditure. To more accurately model user budgets, we also consider a user's within-session budget, which may deviate from their global budget depending on their shopping context. In this paper, we find that users implicitly reveal their session-specific budgets through the sequence of items they browse within that session. Specifically, we find that some users "browse down," by purchasing the cheapest item among alternatives under consideration, others "browse up" (selecting the most expensive), and others ultimately purchase items around the middle. Surprisingly, this mixture of behaviors is difficult to observe globally, as individual users tend to belong firmly to one of the three segments. To model this behavior, we develop an interpretable budget model that combines a clustering component to detect different user segments, with a model of segment-specific purchase profiles. We apply our model on a dataset of browsing and purchasing sessions from Etsy, a large e-commerce website focused on handmade and vintage goods, where it outperforms strong baselines and existing production systems.

User-Generated Content (UGC) is at the core of web applications where users can both produce and consume content. This differs from traditional e-Commerce domains where content producers and consumers are usually from two separate groups. In this work, we propose a method CPRec (consumer and producer based recommendation), for recommending content on UGC-based platforms. Specifically, we learn a core embedding for each user and two transformation matrices to project the user's core embedding into two 'role' embeddings (i.e., a producer and consumer role). We model each interaction by the ternary relation between the consumer, the consumed item, and its producer. Empirical studies on two large-scale UGC applications show that our method outperforms standard collaborative filtering methods as well as recent methods that model producer information via item features.

Product size recommendation and fit prediction are critical in order to improve customers' shopping experiences and to reduce product return rates. Modeling customers' fit feedback is challenging due to its subtle semantics, arising from the subjective evaluation of products, and imbalanced label distribution. In this paper, we propose a new predictive framework to tackle the product fit problem, which captures the semantics behind customers' fit feedback, and employs a metric learning technique to resolve label imbalance issues. We also contribute two public datasets collected from online clothing retailers.

Sequential recommendation algorithms aim to predict users' future behavior given their historical interactions. A recent line of work has achieved state-of-the-art performance on sequential recommendation tasks by adapting ideas from metric learning and knowledge-graph completion. These algorithms replace inner products with low-dimensional embeddings and distance functions, employing a simple translation dynamic to model user behavior over time.

'Explicit' and 'implicit' feedback in recommender systems have been studied for many years, as two relatively isolated areas. However many real-world systems involve a spectrum of both implicit and explicit signals, ranging from clicks and purchases, to ratings and reviews. A natural question is whether implicit signals (which are dense but noisy) might help to predict explicit signals (which are sparse but reliable), or vice versa. Thus in this paper, we propose an item recommendation framework which jointly models this full spectrum of interactions. Our main observation is that in many settings, feedback signals exhibit monotonic dependency structures, i.e., any signal necessarily implies the presence of a weaker (or more implicit) signal (a 'review' action implies a 'purchase' action, which implies a 'click' action, etc.). We refer to these structures as 'monotonic behavior chains,' for which we develop new algorithms that exploit these dependencies. Using several new and existing datasets that exhibit a variety of feedback types, we demonstrate the quantitative performance of our approaches. We also perform qualitative analysis to uncover the relationships between different stages of implicit vs. explicit signals.

Personalized ranking with implicit feedback (e.g. purchases, views, check-ins) is an important paradigm in recommender systems. Such feedback sometimes comes with textual information (e.g. reviews, comments, tips), which could be a useful signal to reveal item properties, identify users' tastes and interpret their behavior. Although incorporating such information is common in explicit feedback settings (such as rating prediction), it is less common when dealing with implicit feedback, as it is often not available for negative instances (e.g. there is no review associated with the item the user didn't buy). Thus our goal in this study is to propose a ranking method (PRAST) to incorporate such personalized, asymmetric textual signals in implicit feedback settings. We evaluate our model on two real-world datasets. Quantitative and qualitative results indicate that the proposed approach significantly outperforms standard recommendation baselines, alleviates ‘cold start’ issues, and is able to provide potential textual interpretations for latent feedback dimensions.

Solving constrained combinatorial optimisation problems via MAP inference is often achieved by introducing extra potential functions for each constraint. This can result in very high order potentials, e.g. a 2nd-order objective with pairwise potentials and a quadratic constraint over all N variables would correspond to an unconstrained objective with an order-N potential. This limits the practicality of such an approach, since inference with high order potentials is tractable only for a few special classes of functions. We propose an approach which is able to solve constrained combinatorial problems using belief propagation without increasing the order. For example, in our scheme the 2nd-order problem above remains order 2 instead of order N. Experiments on applications ranging from foreground detection, image reconstruction, quadratic knapsack, and the M-best solutions problem demonstrate the effectiveness and efficiency of our method. Moreover, we show several situations in which our approach outperforms commercial solvers like CPLEX and others designed for specific constrained MAP inference problems.

Building effective recommender systems for domains like fashion is challenging due to the high level of subjectivity and the semantic complexity of the features involved (i.e., fashion styles). Recent work has shown that approaches to 'visual' recommendation (e.g. clothing, art, etc.) can be made more accurate by incorporating visual signals directly into the recommendation objective, using 'off-the-shelf' feature representations derived from deep networks. Here, we seek to extend this contribution by showing that recommendation performance can be significantly improved by learning 'fashion aware' image representations directly, i.e., by training the image representation (from the pixel level) and the recommender system jointly; this contribution is related to recent work using Siamese CNNs, though we are able to show improvements over state-of-the-art recommendation techniques such as BPR and variants that make use of pretrained visual features. Furthermore, we show that our model can be used generatively, i.e., given a user and a product category, we can generate new images (i.e., clothing items) that are most consistent with their personal taste. This represents a first step towards building systems that go beyond recommending existing items from a product corpus, but which can be used to suggest styles and aid the design of new products.

Dance Dance Revolution (DDR) is a popular rhythm-based video game. Players perform steps on a dance platform in synchronization with music as directed by on-screen step charts. While many step charts are available in standardized packs, players may grow tired of existing charts, or wish to dance to a song for which no chart exists. We introduce the task of learning to choreograph. Given a raw audio track, the goal is to produce a new step chart. This task decomposes naturally into two subtasks: deciding when to place steps and deciding which steps to select. For the step placement task, we combine recurrent and convolutional neural networks to ingest spectrograms of low-level audio features to predict steps, conditioned on chart difficulty. For step selection, we present a conditional LSTM generative model that substantially outperforms n-gram and fixed-window approaches.

Dealing with sparse, long-tailed datasets, and cold-start problems is always a challenge for recommender systems. These issues can partly be dealt with by making predictions not in isolation, but by leveraging information from related events; such information could include signals from social relationships or from the sequence of recent activities. Both types of additional information can be used to improve the performance of state-of-the-art matrix factorization-based techniques. In this paper, we propose new methods to combine both social and sequential information simultaneously, in order to further improve recommendation performance. We show these techniques to be particularly effective when dealing with sparsity and cold-start issues in several large, real-world datasets.

Hyper graph matching problems have drawn attention recently due to their ability to embed higher order relations between nodes. In this paper, we formulate hyper graph matching problems as constrained MAP inference problems in graphical models. Whereas previous discrete approaches introduce several global correspondence vectors, we introduce only one global correspondence vector, but several local correspondence vectors. This allows us to decompose the problem into a (linear) bipartite matching problem and several belief propagation sub-problems. Bipartite matching can be solved by traditional approaches, while the belief propagation sub-problem is further decomposed as two sub-problems with optimal substructure. Then a newly proposed dynamic programming procedure is used to solve the belief propagation sub-problem. Experiments show that the proposed methods outperform state-of-the-art techniques for hyper graph matching.

Traditional approaches to recommendation focus on learning from large volumes of historical feedback to estimate simple numerical quantities (Will a user click on a product? Make a purchase? etc.). Natural language approaches that model information like product reviews have proved to be incredibly useful in improving the performance of such methods, as reviews provide valuable auxiliary information that can be used to better estimate latent user preferences and item properties. In this paper, rather than using reviews as an inputs to a recommender system, we focus on generating reviews as the model’s output. This requires us to efficiently model text (at the character level) to capture the preferences of the user, the properties of the item being consumed, and the interaction between them (i.e., the user’s preference). We show that this can model can be used to (a) generate plausible reviews and estimate nuanced reactions; (b) provide personalized rankings of existing reviews; and (c) recommend existing products more effectively.

Recommender systems are ubiquitous in applications ranging from e-commerce to social media, helping users to navigate a huge selection of items and to meet a variety of special needs and user tastes. Incorporating contextual knowledge into such systems - such as relational information - has proven to be an effective way to improve recommendation accuracy. A popular line of research aims to model relationships between users, through their connections in a social network. In contrast, we aim to model complex relationships between products, using data based on co-purchase and co-browsing behavior. Modeling such networks presents a variety of challenges, in particular because the features that make two items complementary (or likely to be co-purchased) are far more complex than mere similarity. To model these complex relationships we develop a method based on pairwise ranking and embedding learning to build representations of items based on their co-purchasing and co-browsing statistics. We conduct experiments on Amazon product data to demonstrate that modeling such relationships significantly improves accuracy compared to competitive baselines.

Modeling the complex interactions between users and items as well as amongst items themselves is at the core of designing successful recommender systems. One classical setting is predicting users' personalized sequential behavior (or 'next-item' recommendation), where the challenges mainly lie in modeling 'third-order' interactions between a user, her previously visited item(s), and the next item to consume. Existing methods typically decompose these higher-order interactions into a combination of pairwise relationships, by way of which user preferences (user-item interactions) and sequential patterns (item-item interactions) are captured by separate components. In this paper, we propose a unified method, TransRec, to model such third-order relationships for large-scale sequential prediction. Methodologically, we embed items into a 'transition space' where users are modeled as translation vectors operating on item sequences. Empirically, this approach outperforms the state-of-the-art on a wide spectrum of real-world datasets. Data and code are available at https://sites.google.com/a/eng.ucsd.edu/ruining-he/.

Many websites offer promotions in terms of bundled items that can be purchased together, usually at a discounted rate. 'Bundling' may be a means of increasing sales revenue, but may also be a means for content creators to expose users to new items that they may not have considered in isolation. In this paper, we seek to understand the semantics of what constitutes a 'good' bundle, in order to recommend existing bundles to users on the basis of their constituent products, as well the more difficult task of generating new bundles that are personalized to a user. To do so we collect a new dataset from the Steam video game distribution platform, which is unique in that it contains both 'traditional' recommendation data (rating and purchase histories between users and items), as well as bundle purchase information. We assess issues such as bundle size and item compatibility, and show that these features, when combined with traditional matrix factorization techniques, can lead to highly effective bundle recommendation and generation.

Bartering is a timeless practice that is becoming increasingly popular on the Web. Recommending trades for an online bartering platform shares many similarities with traditional approaches to recommendation, in particular the need to model the preferences of users and the properties of the items they consume. However, there are several aspects that make bartering problems interesting and challenging, specifically the fact that users are both suppliers and consumers, and that the trading environment is highly dynamic. Thus, a successful model of bartering requires us to understand not just users' preferences, but also the social dynamics of who trades with whom, and the temporal dynamics of when trades occur.

In order to match shoppers with desired products and provide personalized promotions, whether in online or offline shopping worlds, it is critical to model both consumer preferences and price sensitivities simultaneously. Personalized preferences have been thoroughly studied in the field of recommender systems, though price (and price sensitivity) has received relatively little attention. At the same time, price sensitivity has been richly explored in the area of economics, though typically not in the context of developing scalable, working systems to generate recommendations. In this study, we seek to bridge the gap between large-scale recommender systems and established consumer theories from economics, and propose a nested feature-based matrix factorization framework to model both preferences and price sensitivities. Quantitative and qualitative results indicate the proposed personalized, interpretable and scalable framework is capable of providing satisfying recommendations (on two datasets of grocery transactions) and can be applied to obtain economic insights into consumer behavior.

Modern recommender systems model people and items by discovering or `teasing apart' the underlying dimensions that encode the properties of items and users' preferences toward them. Critically, such dimensions are uncovered based on user feedback, often in implicit form (such as purchase histories, browsing logs, etc.); in addition, some recommender systems make use of side information, such as product attributes, temporal information, or review text.However one important feature that is typically ignored by existing personalized recommendation and ranking methods is the visual appearance of the items being considered. In this paper we propose a scalable factorization model to incorporate visual signals into predictors of people's opinions, which we apply to a selection of large, real-world datasets. We make use of visual features extracted from product images using (pre-trained) deep networks, on top of which we learn an additional layer that uncovers the visual dimensions that best explain the variation in people's feedback. This not only leads to significantly more accurate personalized ranking methods, but also helps to alleviate cold start issues, and qualitatively to analyze the visual dimensions that influence people's opinions.

Feature matching is a key problem in computer vision and pattern recognition. One way to encode the essential interdependence between potential feature matches is to cast the problem as inference in a graphical model, though recently alternatives such as spectral methods, or approaches based on the convex-concave procedure have achieved the state-of-the-art. Here we revisit the use of graphical models for feature matching, and propose a belief propagation scheme which exhibits the following advantages: (1) we explicitly enforce one-to-one matching constraints, (2) we offer a tighter relaxation of the original cost function than previous graphical-model-based approaches, and (3) our sub-problems decompose into max-weight bipartite matching, which can be solved efficiently, leading to orders-of-magnitude reductions in execution time. Experimental results show that the proposed algorithm produces results superior to those of the current state-of-the-art.

Predicting personalized sequential behavior is a key task for recommender systems. In order to predict user actions such as the next product to purchase, movie to watch, or place to visit, it is essential to take into account both long-term user preferences and sequential patterns (i.e., short-term dynamics). Matrix Factorization and Markov Chain methods have emerged as two separate but powerful paradigms for modeling the two respectively. Combining these ideas has led to unified methods that accommodate long-and short-term dynamics simultaneously by modeling pairwise user-item and item-item interactions. In spite of the success of such methods for tackling dense data, they are challenged by sparsity issues, which are prevalent in real-world datasets. In recent years, similarity-based methods have been proposed for (sequentially-unaware) item recommendation with promising results on sparse datasets. In this paper, we propose to fuse such methods with Markov Chains to make personalized sequential recommendations. We evaluate our method, Fossil, on a variety of large, real-world datasets. We show quantitatively that Fossil outperforms alternative algorithms, especially on sparse datasets, and qualitatively that it captures personalized dynamics and is able to make meaningful recommendations.

Identifying relationships between items is a key task of an online recommender system, in order to help users discover items that are functionally complementary or visually compatible. In domains like clothing recommendation, this task is particularly challenging since a successful system should be capable of handling a large corpus of items, a huge amount of relationships among them, as well as the high-dimensional and semantically complicated features involved. Furthermore, the human notion of "compatibility" to capture goes beyond mere similarity: For two items to be compatible-whether jeans and a t-shirt, or a laptop and a charger-they should be similar in some ways, but systematically different in others. In this paper we propose a novel method, Monomer, to learn complicated and heterogeneous relationships between items in product recommendation settings. Recently, scalable methods have been developed that address this task by learning similarity metrics on top of the content of the products involved. Here our method relaxes the metricity assumption inherent in previous work and models multiple localized notions of 'relatedness,' so as to uncover ways in which related items should be systematically similar, and systematically different. Quantitatively, we show that our system achieves state-of-the-art performance on large-scale compatibility prediction tasks, especially in cases where there is substantial heterogeneity between related items.

Product review websites provide an incredible lens into the wide variety of opinions and experiences of different people, and play a critical role in helping users discover products that match their personal needs and preferences. To help address questions that can't easily be answered by reading others' reviews, some review websites also allow users to pose questions to the community via a question-answering (QA) system. As one would expect, just as opinions diverge among different reviewers, answers to such questions may also be subjective, opinionated, and divergent. This means that answering such questions automatically is quite different from traditional QA tasks, where it is assumed that a single 'correct' answer is available. While recent work introduced the idea of question-answering using product reviews, it did not account for two aspects that we consider in this paper: (1) Questions have multiple, often divergent, answers, and this full spectrum of answers should somehow be used to train the system, and (2) What makes a 'good' answer depends on the asker and the answerer, and these factors should be incorporated in order for the system to be more personalized. Here we build a new QA dataset with 800 thousand questions-and over 3.1 million answers-and show that explicitly accounting for personalization and ambiguity leads both to quantitatively better answers, but also a more nuanced view of the range of supporting, but subjective, opinions.

Understanding users' interactions with highly subjective content---like artistic images---is challenging due to the complex semantics that guide our preferences. On the one hand one has to overcome `standard' recommender systems challenges, such as dealing with large, sparse, and long-tailed datasets. On the other, several new challenges present themselves, such as the need to model content in terms of its visual appearance, or even social dynamics, such as a preference toward a particular artist that is independent of the art they create. In this paper we build large-scale recommender systems to model the dynamics of a vibrant digital art community, Behance, consisting of tens of millions of interactions (clicks and 'appreciates') of users toward digital art. Methodologically, our main contributions are to model (a) rich content, especially in terms of its visual appearance; (b) temporal dynamics, in terms of how users prefer 'visually consistent' content within and across sessions; and (c) social dynamics, in terms of how users exhibit preferences both towards certain art styles, as well as the artists themselves.

Building a successful recommender system depends on understanding both the dimensions of people's preferences as well as their dynamics. In certain domains, such as fashion, modeling such preferences can be incredibly difficult, due to the need to simultaneously model the visual appearance of products as well as their evolution over time. The subtle semantics and non-linear dynamics of fashion evolution raise unique challenges especially considering the sparsity and large scale of the underlying datasets. In this paper we build novel models for the One-Class Collaborative Filtering setting, where our goal is to estimate users' fashion-aware personalized ranking functions based on their past feedback. To uncover the complex and evolving visual factors that people consider when evaluating products, our method combines high-level visual features extracted from a deep convolutional neural network, users' past feedback, as well as evolving trends within the community. Experimentally we evaluate our method on two large real-world datasets from Amazon.com, where we show it to outperform state-of-the-art personalized ranking measures, and also use it to visualize the high-level fashion trends across the 11-year span of our dataset.

Online reviews are often our first port of call when considering products and purchases online. When evaluating a potential purchase, we may have a specific query in mind, e.g. `will this baby seat fit in the overhead compartment of a 747?' or `will I like this album if I liked Taylor Swift's 1989?'. To answer such questions we must either wade through huge volumes of consumer reviews hoping to find one that is relevant, or otherwise pose our question directly to the community via a Q/A system.

Latent Factor models, which transform both users and items into the same latent feature space, are one of the most successful and ubiquitous models in recommender systems. Most existing models in this paradigm define both users' and items' latent factors to be of the same size and use an inner product to represent a user's "compatibility" with an item. Intuitively, users' factors encode "preferences" while item factors encode "properties", so that the inner product encodes how well an item matches a user's preferences. However, a user's opinion of an item may be more complex, for example each dimension of each user's opinion may depend on a combination of multiple item factors simultaneously. Thus it may be better to view each dimension of a user's preference as a personalized projection of an item's properties so that the preference model can capture complex relationships between items' properties and users' preferences.

With the rapid proliferation of smart mobile devices, users now take millions of photos every day. These include large numbers of clothing and accessory images. We would like to answer questions like 'What outfit goes well with this pair of shoes?' To answer these types of questions, one has to go beyond learning visual similarity and learn a visual notion of compatibility across categories. In this paper, we propose a novel learning framework to help answer these types of questions. The main idea of this framework is to learn a feature transformation from images of items into a latent space that expresses compatibility. For the feature transformation, we use a Siamese Convolutional Neural Network (CNN) architecture, where training examples are pairs of items that are either compatible or incompatible. We model compatibility based on co-occurrence in large-scale user behavior data, in particular co-purchase data from Amazon.com. To learn cross-category fit, we introduce a strategic method to sample training data, where pairs of items are heterogeneous dyads, i.e., the two elements of a pair belong to different high-level categories. While this approach is applicable to a wide variety of settings, we focus on the representative problem of learning compatible clothing style. Our results indicate that the proposed framework is capable of learning semantic information about visual style and is able to generate outfits of clothes, with items from different categories, that go well together.

To design a useful recommender system, it is important to understand how products relate to each other. For example, while a user is browsing mobile phones, it might make sense to recommend other phones, but once they buy a phone, we might instead want to recommend batteries, cases, or chargers. In economics, these two types of recommendations are referred to as substitutes and complements: substitutes are products that can be purchased instead of each other, while complements are products that can be purchased in addition to each other. Such relationships are essential as they help us to identify items that are relevant to a user's search.

Humans inevitably develop a sense of the relationships between objects, some of which are based on their appearance. Some pairs of objects might be seen as being alternatives to each other (such as two pairs of jeans), while others may be seen as being complementary (such as a pair of jeans and a matching shirt). This information guides many of the choices that people make, from buying clothes to their interactions with each other. We seek here to model this human sense of the relationships between objects based on their appearance. Our approach is not based on fine-grained modeling of user annotations but rather on capturing the largest dataset possible and developing a scalable method for uncovering human notions of the visual relationships within. We cast this as a network inference problem defined on graphs of related images, and provide a large-scale dataset for the training and evaluation of the same. The system we develop is capable of recommending which clothes and accessories will go well together (and which will not), amongst a host of other applications.

Recommending products to users means estimating their preferences for certain items over others. This can be cast either as a problem of estimating the rating that each user will give to each item, or as a problem of estimating users' relative preferences in the form of a ranking. Although collaborative-filtering approaches can be used to identify users who rate and rank products similarly, another source of data that informs us about users' preferences is their set of social connections. Both rating- and ranking-based paradigms are important in real-world recommendation settings, though rankings are especially important in settings where explicit feedback in the form of a numerical rating may not be available. Although many existing works have studied how social connections can be used to build better models for rating prediction, few have used social connections as a means to derive more accurate ranking-based models. Using social connections to better estimate users' rankings of products is the task we consider in this paper. We develop a model, SBPR (Social Bayesian Personalized Ranking), based on the simple observation that users tend to assign higher ranks to items that their friends prefer. We perform experiments on four real-world recommendation data sets, and show that SBPR outperforms alternatives in ranking prediction both in warm- and cold-start settings.

Networks are a general language for representing relational information among objects. An effective way to model, reason about, and summarize networks, is to discover sets of nodes with common connectivity patterns. Such sets are commonly referred to as network communities. Research on network community detection has predominantly focused on identifying communities of densely connected nodes in undirected networks.

Event sequences, such as patients' medical histories or users' sequences of product reviews, trace how individuals progress over time. Identifying common patterns, or progression stages, in such event sequences is a challenging task because not every individual follows the same evolutionary pattern, stages may have very different lengths, and individuals may progress at different rates. In this paper, we develop a model-based method for discovering common progression stages in general event sequences. We develop a generative model in which each sequence belongs to a class, and sequences from a given class pass through a common set of stages, where each sequence evolves at its own rate. We then develop a scalable algorithm to infer classes of sequences, while also segmenting each sequence into a set of stages. We evaluate our method on event sequences, ranging from patients' medical histories to online news and navigational traces from the Web. The evaluation shows that our methodology can predict future events in a sequence, while also accurately inferring meaningful progression stages, and effectively grouping sequences based on common progression patterns. More generally, our methodology allows us to reason about how event sequences progress over time, by discovering patterns and categories of temporal evolution in large-scale datasets of events.

People's personal social networks are big and cluttered, and currently there is no good way to automatically organize them. Social networking sites allow users to manually categorize their friends into social circles (e.g., “circles” on Google+, and “lists” on Facebook and Twitter). However, circles are laborious to construct and must be manually updated whenever a user's network grows. In this article, we study the novel task of automatically identifying users' social circles. We pose this task as a multimembership node clustering problem on a user's ego network, a network of connections between her friends. We develop a model for detecting circles that combines network structure as well as user profile information. For each circle, we learn its members and the circle-specific user profile similarity metric. Modeling node membership to multiple circles allows us to detect overlapping as well as hierarchically nested circles. Experiments show that our model accurately identifies circles on a diverse set of data from Facebook, Google+, and Twitter, for all of which we obtain hand-labeled ground truth.

Community detection algorithms are fundamental tools that allow us to uncover organizational principles in networks. When detecting communities, there are two possible sources of information one can use: the network structure, and the features and attributes of nodes. Even though communities form around nodes that have common edges and common attributes, typically, algorithms have only focused on one of these two data modalities: community detection algorithms traditionally focus only on the network structure, while clustering algorithms mostly consider only node attributes. In this paper, we develop Communities from Edge Structure and Node Attributes (CESNA), an accurate and scalable algorithm for detecting overlapping communities in networks with node attributes. CESNA statistically models the interaction between the network structure and the node attributes, which leads to more accurate community detection as well as improved robustness in the presence of noise in the network structure. CESNA has a linear runtime in the network size and is able to process networks an order of magnitude larger than comparable approaches. Last, CESNA also helps with the interpretation of detected communities by finding relevant node attributes for each community.

In order to recommend products to users we must ultimately predict how a user will respond to a new product. To do so we must uncover the implicit tastes of each user as well as the properties of each product. For example, in order to predict whether a user will enjoy Harry Potter, it helps to identify that the book is about wizards, as well as the user's level of interest in wizardry. User feedback is required to discover these latent product and user dimensions. Such feedback often comes in the form of a numeric rating accompanied by review text. However, traditional methods often discard review text, which makes user and product latent dimensions difficult to interpret, since they ignore the very text that justifies a user's rating. In this paper, we aim to combine latent rating dimensions (such as those of latent-factor recommender systems) with latent review topics (such as those learned by topic models like LDA). Our approach has several advantages. Firstly, we obtain highly interpretable textual labels for latent rating dimensions, which helps us to `justify' ratings with text. Secondly, our approach more accurately predicts product ratings by harnessing the information present in review text; this is especially true for new products and users, who may have too few ratings to model their latent factors, yet may still provide substantial information from the text of even a single review. Thirdly, our discovered topics can be used to facilitate other tasks such as automated genre discovery, and to identify useful and representative reviews.

Recommending products to consumers means not only understanding their tastes, but also understanding their level of experience. For example, it would be a mistake to recommend the iconic film Seven Samurai simply because a user enjoys other action movies; rather, we might conclude that they will eventually enjoy it---once they are ready. The same is true for beers, wines, gourmet foods---or any products where users have acquired tastes: the `best' products may not be the most 'accessible'. Thus our goal in this paper is to recommend products that a user will enjoy now, while acknowledging that their tastes may have changed over time, and may change again in the future. We model how tastes change due to the very act of consuming more products---in other words, as users become more experienced. We develop a latent factor recommendation system that explicitly accounts for each user's level of experience. We find that such a model not only leads to better recommendations, but also allows us to study the role of user experience and expertise on a novel dataset of fifteen million beer, wine, food, and movie reviews.

Most online reviews consist of plain-text feedback together with a single numeric score. However, understanding the multiple `aspects' that contribute to users' ratings may help us to better understand their individual preferences. For example, a user's impression of an audio book presumably depends on aspects such as the story and the narrator, and knowing their opinions on these aspects may help us to recommend better products. In this paper, we build models for rating systems in which such dimensions are explicit, in the sense that users leave separate ratings for each aspect of a product. By introducing new corpora consisting of five million reviews, rated with between three and six aspects, we evaluate our models on three prediction tasks: First, we uncover which parts of a review discuss which of the rated aspects. Second, we summarize reviews by finding the sentences that best explain a user's rating. Finally, since aspect ratings are optional in many of the datasets we consider, we recover ratings that are missing from a user's evaluation. Our model matches state-of-the-art approaches on existing small-scale datasets, while scaling to the real-world datasets we introduce. Moreover, our model is able to `disentangle' content and sentiment words: we automatically learn content words that are indicative of a particular aspect as well as the aspect-specific sentiment words that are indicative of a particular rating.

Our personal social networks are big and cluttered, and currently there is no good way to organize them. Social networking sites allow users to manually categorize their friends into social circles (e.g. circles' on Google+, andlists' on Facebook and Twitter), however they are laborious to construct and must be updated whenever a user's network grows. We define a novel machine learning task of identifying users' social circles. We pose the problem as a node clustering problem on a user's ego-network, a network of connections between her friends. We develop a model for detecting circles that combines network structure as well as user profile information. For each circle we learn its members and the circle-specific user profile similarity metric. Modeling node membership to multiple circles allows us to detect overlapping as well as hierarchically nested circles. Experiments show that our model accurately identifies circles on a diverse set of data from Facebook, Google+, and Twitter for all of which we obtain hand-labeled ground-truth data.

The MAP inference problem in many graphical models can be solved efficiently using a fast algorithm for computing min-sum products of n × n matrices. The class of models in question includes cyclic and skip-chain models that arise in many applications. Although the worst-case complexity of the min-sum product operation is not known to be much better than O(n(3)), an O(n(2.5)) expected time algorithm was recently given, subject to some constraints on the input matrices. In this paper, we give an algorithm that runs in O(n(2) log n) expected time, assuming that the entries in the input matrices are independent samples from a uniform distribution. We also show that two variants of our algorithm are quite fast for inputs that arise in several applications. This leads to significant performance gains over previous methods in applications within computer vision and natural language processing.

Graph matching is a classical problem in pattern recognition with many applications, particularly when the graphs are embedded in Euclidean spaces, as is often the case for computer vision. There are several variants of the matching problem, concerned with isometries, isomorphisms, homeomorphisms, and node attributes; different approaches exist for each variant. We show how structured estimation methods from machine learning can be used to combine such variants into a single version of graph matching. In this paradigm, the extent to which our datasets reveal isometries, isomorphisms, homeomorphisms, and other properties is automatically accounted for in the learning process so that any such specific qualification of graph matching loses meaning. We present experiments with real computer vision data showing the leverage of this unified formulation.

We present a graphical model that encodes hierarchical constraints for classifying image regions at multiple scales. We show that inference can be performed efficiently and exactly, rendering it amenable to structured learning. Our model is parametrised using the outputs of a series of first-order classifiers, meaning that it learns which classifiers are useful at different scales, as well as the relationships between classifiers across scales. Example results Correct labeling, using bounding-boxes from VOC2007 (1 − ∆ = 1): Baseline, using no second-order features (1 − ∆ = 0.566): Our model The ‘nodes ’ of our graphical model correspond to overlapping image regions: Second-order features without learning (1 − ∆ = 0.551): Learning of all features (1 − ∆ = 0.770): Colour-code for labels: Edges are formed by connecting nodes at different scales: we connect two nodes precisely when the corresponding image regions overlap at adjacent scales, so that our graphical model forms a quad-tree. First-order (node) features Our image features are based on those from [2], in which image-level, region-level, and patch-level classifiers are proposed. We use all classifiers at all scales (Pr,label is the probability that the region r is labeled label): Φ nodes (r, label) = (0,..., P 1 r,label,..., 0)... (0,..., P features for first classifier n r,label,..., 0) features for nth classifier Thus we learn which classifiers are useful at which scales. Hierarchical constraints We want to ban inconsistent assignments at different scales: aeroplane

Many traditional methods for shape classification involve establishing point correspondences between shapes to produce matching scores, which are in turn used as similarity measures for classification. Learning techniques have been applied only in the second stage of this process, after the matching scores have been obtained. In this paper, instead of simply taking for granted the scores obtained by matching and then learning a classifier, we learn the matching scores themselves so as to produce shape similarity scores that minimize the classification loss. The solution is based on a max-margin formulation in the structured prediction setting. Experiments in shape databases reveal that such an integrated learning algorithm substantially improves on existing methods.

We present a method for learning max-weight matching predictors in bipartite graphs. The method consists of performing maximum a posteriori estimation in exponential families with sufficient statistics that encode permutations and data features. Although inference is in general hard, we show that for one very relevant application - document ranking - exact inference is efficient. For general model instances, an appropriate sampler is readily available. Contrary to existing max-margin matching models, our approach is statistically consistent and, in addition, experiments with increasing sample sizes indicate superior improvement over such models. We apply the method to graph matching in computer vision as well as to a standard benchmark dataset for learning document ranking, in which we obtain state-of-the-art results, in particular improving on max-margin variants. The drawback of this method with respect to max-margin alternatives is its runtime for large graphs, which is high comparatively.

As a fundamental problem in pattern recognition, graph matching has applications in a variety of fields, from computer vision to computational biology. In graph matching, patterns are modeled as graphs and pattern recognition amounts to finding a correspondence between the nodes of different graphs. Many formulations of this problem can be cast in general as a quadratic assignment problem, where a linear term in the objective function encodes node compatibility and a quadratic term encodes edge compatibility. The main research focus in this theme is about designing efficient algorithms for approximately solving the quadratic assignment problem, since it is NP-hard. In this paper we turn our attention to a different question: how to estimate compatibility functions such that the solution of the resulting graph matching problem best matches the expected solution that a human would manually provide. We present a method for learning graph matching: the training examples are pairs of graphs and the 'labels' are matches between them. Our experimental results reveal that learning can substantially improve the performance of standard graph matching algorithms. In particular, we find that simple linear assignment with such a learning scheme outperforms Graduated Assignment with bistochastic normalisation, a state-of-the-art quadratic assignment relaxation algorithm.

Models for near-rigid shape matching are typically based on distance-related features, in order to infer matches that are consistent with the isometric assumption. However, real shapes from image datasets, even when expected to be related by almost isometric" transformations, are actually subject not only to noise but also, to some limited degree, to variations in appearance and scale. In this paper, we introduce a graphical model that parameterises appearance, distance, and angle features and we learn all of the involved parameters via structured prediction. The outcome is a model for near-rigid shape matching which is robust in the sense that it is able to capture the possibly limited but still important scale and appearance variations. Our experimental results reveal substantial improvements upon recent successful models, while maintaining similar running times."

A recent paper [1] proposed a provably optimal polynomial time method for performing near-isometric point pattern matching by means of exact probabilistic inference in a chordal graphical model. Its fundamental result is that the chordal graph in question is shown to be globally rigid, implying that exact inference provides the same matching solution as exact inference in a complete graphical model. This implies that the algorithm is optimal when there is no noise in the point patterns. In this paper, we present a new graph that is also globally rigid but has an advantage over the graph proposed in [1]: Its maximal clique size is smaller, rendering inference significantly more efficient. However, this graph is not chordal, and thus, standard Junction Tree algorithms cannot be directly applied. Nevertheless, we show that loopy belief propagation in such a graph converges to the optimal solution. This allows us to retain the optimality guarantee in the noiseless case, while substantially reducing both memory requirements and processing time. Our experimental results show that the accuracy of the proposed solution is indistinguishable from that in [1] when there is noise in the point patterns.

In this paper, we use large neighborhood Markov random fields to learn rich prior models of color images. Our approach extends the monochromatic Fields of Experts model (Roth & Black, 2005a) to color images. In the Fields of Experts model, the curse of dimensionality due to very large clique sizes is circumvented by parameterizing the potential functions according to a product of experts. We introduce simplifications to the original approach by Roth and Black which allow us to cope with the increased clique size (typically 3x3x3 or 5x5x3 pixels) of color images. Experimental results are presented for image denoising which evidence improvements over state-of-the-art monochromatic image priors.