Multimodal Large Language Models (MLLMs) have recently achieved strong performance across a variety of multimodal tasks. However, they still suffer from various forms of hallucination, which hinder their practical deployment. Prior approaches often struggle to efficiently construct high-quality hallucination-related samples and to process them in a fine-grained manner, resulting in limited effectiveness in hallucination alleviation. To address this issue, we propose a data sampling strategy that selects samples better suited for hallucination-oriented training, thereby enhancing training effectiveness. In addition, we introduce a quantitative method for measuring hallucination severity and assign individualized weights to training samples accordingly. Building on this, we present Hallucination-Differentiated Direct Preference Optimization (HD-DPO), a novel preference optimization framework. During fine-tuning, HD-DPO incorporates these weights into both the formulation of customized loss functions and the modulation of localized visual attention, enabling fine-grained optimization. Experimental results demonstrate that our method outperforms existing fine-tuning strategies across multiple benchmarks and generalizes well to diverse MLLM architectures, effectively reducing hallucination rates and enhancing overall model performance.

Ensuring consistent persona in interactive AI systems presents a significant challenge, especially in diverse application scenarios ranging from virtual assistants to customer service bots. Such capability is often constrained by the system's understanding of direct and explicit persona conflicts. Traditional approaches primarily focus on detecting discrepancies between machine responses and its predefined profile, or the contextual inconsistencies between the responses at the semantic level rather than the persona level. Due to the lack of a comprehensive persona-specific Commonsense Knowledge Graph, some indirect and implicit persona inconsistencies between machine responses can hardly be identified. In this paper, we build the first persona commonsense knowledge graph (PersonaKG), based on which we then construct a large-scale persona consistency dialogue dataset (PersonaCOM) containing both explicit and implicit persona conflicts between machine responses. With the guidance of the persona commonsense knowledge, we propose a Recognize-Rewrite framework (R2) which first recognizes the responses that are inconsistent in persona with the previous responses, and then rewrites them into consistent ones. The empirical study demonstrates that utilizing R2 method on PersonaCOM with PersonaKG results in a significant improvement of 12.20% in automatic metrics and 10.09% in manual evaluation compared to not using the R2 method and PersonaKG.

Evaluating the coding capabilities of models through algorithmic code generation is challenging, as it requires deep problem understanding and complex algorithm design. Current benchmarks suffer from a narrow focus on final execution results (such as pass@k), neglecting the crucial reasoning and problem-solving processes inherent in code generation. To address this limitation, we introduce a multi-phase algorithmic code generation benchmark, MUPA, structured around human computational thinking. MUPA dissects the evaluation into four distinct phases: example understanding, algorithm selection, solution description, and code generation. This framework facilitates a comprehensive assessment by providing insights into the model's intermediate problem-solving steps, rather than just the final code. We manually curated 197 high-quality competitive programming problems from Codeforces. Utilizing an LLM-as-a-judge paradigm with specialized prompts, our rigorous evaluation of several existing code generation LLMs reveals significant across-the-board challenges. Notably, we establish a positive correlation, indicating that proficiency in an earlier phase directly impacts performance in subsequent phases, underscoring the interdependency of these algorithmic skills. The benchmark is publicly available at https://github.com/cheniison/MUPA.

Self-training for Named Entity Recognition (NER) aims at identifying named entities and their types in the text using self-training to fully make use of the limited labeled data and a large amount of unlabeled data. The major challenge in self-training is confirmation bias where incorrect pseudo-labels increase errors. Many efforts have been made to address this challenge, but few labeled data limit their performance. In this paper, we introduce Large Language Model (LLM) into self-training to select high-quality pseudo-labels leveraging its rich knowledge and few-shot learning capability. Specifically, we design a comprehensive prompt to improve the judgment performance of LLM, where the prompt incorporates task rules mined by LLM itself to fully leverage labeled data. In addition, to reduce the impact of LLM's hallucinations, we adopt a collaborative pseudo-label selection based on combined confidence and calibration-guided probability smoothing. Our empirical study conducted on several NER datasets shows that our method outperforms state-of-the-art approaches. The code is available at https://github.com/cheniison/llm-judged-ST.

Open Knowledge Bases (Open KBs) are fundamental to knowledge-driven applications, including semantic search, knowledge reasoning, and recommendation systems. However, the presence of redundant and ambiguous expressions within Open KBs significantly hinders their application. This highlights the urgent need for Open KB canonicalization, particularly of attribute values, which comprise nearly 40% of the facts within Open KBs. Unlike entities and predicates, attribute values are inherently sparse and diverse, posing unique challenges for their canonicalization. However, existing studies mainly focus on entities or predicates, leaving attribute value-level noun phrase canonicalization (NPC-AV) underexplored. Large language models (LLMs), with their strengths in common-sense reasoning and fault tolerance, have shown promise in Open KB canonicalization. Yet, current LLM-based approaches often rely heavily on LLM responses, overlooking their high computational cost and potential errors. In this paper, we introduce Caf4AVC, a collaborative framework that integrates clustering-based methods and LLMs for the NPC-AV task. We further propose an innovative two-factor authentication correction mechanism and an adaptive threshold-based selection strategy to address these limitations. Extensive experiments on multiple real-world Open KB datasets demonstrate the effectiveness of our framework, achieving a 17.52% reduction in LLM call costs and a 6.3% average performance improvement compared to competitive methods. The code and dataset are available athttps://github.com/hedyHe/Caf4AV.

While fine-tuned large language models (LLMs) excel in generating grammatically valid SQL in Text-to-SQL parsing, they often struggle to ensure semantic accuracy in queries, leading to user confusion and diminished system usability. To tackle this challenge, we introduce SQLFixAgent, a new consistency-enhanced multi-agent collaborative framework designed for detecting and repairing erroneous SQL. Our framework comprises a core agent, SQLRefiner, alongside two auxiliary agents: SQLReviewer and QueryCrafter. The SQLReviewer agent employs the rubber duck debugging method to identify potential semantic mismatches between SQL and user query. If the error is detected, the QueryCrafter agent generates multiple SQL as candidate repairs using a fine-tuned SQLTool. Subsequently, leveraging similar repair retrieval and failure memory reflection, the SQLRefiner agent selects the most fitting SQL statement from the candidates as the final repair. We evaluated our proposed framework on five Text-to-SQL benchmarks. The experimental results show that our method consistently enhances the performance of the baseline model, specifically achieving an execution accuracy improvement of over 3% on the Bird benchmark. Our framework also has a higher token efficiency compared to other advanced methods, making it more competitive.

Next POI recommendation aids users in predicting their destinations of interest and plays an increasingly vital role in location-based social services. Recent works focus on analyzing both long-term and short-term interests in POI recommendation to gain a deeper understanding of user profiles. However, these methods for modeling long-term user’s sequences primarily rely on the Transformer model, which functions as a low-pass filter, often leading to the loss of high-frequency information. Additionally, long-term and short-term sequences are typically modeled independently, with short-term sequences often defined solely by the most recent check-ins, overlooking their interactions and dependencies. Therefore, we propose Enhancing Long-and Short-Term Representations for Next POI Recommendations via Frequency and Hierarchical Contrastive Learning (FHCRec). FHCRec captures both high-frequency and low-frequency information in long-term sequences to model richer long-term user’s preference representations. Moreover, it harnesses the characteristics of the short-term subsequences embedded within long-term sequences to enhance short-term preference characterization via local and global hierarchical contrastive learning, resulting in more personalized short-term preferences. The enhanced long-term and short-term preferences are integrated to improve model recommendation performance. Extensive experiments on three real-world datasets demonstrate the effectiveness of our method.

Information extraction is a critical technology for intelligent analysis and risk assessment in the dairy financial domain. However, real-world applications face three major challenges: the complexity and diversity of entity-relation types, significant data imbalance, and ambiguity in textual expressions. Traditional methods often fail to capture rare patterns, struggle with vague mentions, and exhibit poor generalization in low-resource settings. To address these issues, we propose a novel framework that integrates large language models (LLMs) with targeted data augmentation and agent-based retrieval-augmented generation (RAG). Our approach builds on the BaiChuan2 model, which is first adapted to the dairy finance domain via secondary pretraining. We introduce a two-stage data augmentation strategy: the first stage uses ChatGPT to generate pseudo-samples for rare types, and the second stage refines model weaknesses based on prediction-guided feedback. These augmented datasets are used to fine-tune the model through prompt-based supervised learning with LoRA. To further enhance robustness, we incorporate an agent-based RAG module for completing vague or underspecified entities by retrieving external contextual knowledge. Extensive experiments demonstrate that our framework achieves state-of-the-art performance, with the improved metric, i.e., F1+ scores, of 0.876 and 0.824 for entity recognition and relation extraction, respectively. The RAG component boosts entity completion accuracy to 0.802 while reducing retrieval latency by over 6x, showcasing both the effectiveness and practicality of our method in real-world dairy financial applications.

Large Language Models (LLMs) exhibit remarkable capabilities across various downstream tasks, including empathetic dialogues. However, a non-trivial question arises: Do they possess high-context empathy and can they generate emotional interactions with humans? High-context empathy, which tends to be more indirect and concise like Chinese-style empathy, differs from the current empathy capabilities of LLMs. These capabilities are predominantly low-context empathy, which is often direct and lengthy, resembling English-style empathy. In this paper, We first construct a comprehensive Chinese High-context Empathy Dialogue dataset (HED), which consists of emotional, role-based emotional, personality-based emotional, and role-personality-based emotional dialogues. Next, we explore whether LLMs have high-context empathy in conversations. After that, we propose an innovative High-context Empathy Network (HEN) to improve LLMs' capabilities in generating high-context empathetic responses. Our empirical study demonstrates that there is much room for LLMs in generating high-context empathetic responses, and the proposed HEN can not only significantly improve LLMs' capabilities in generating high-context empathetic responses, but also has positive effects for LLMs in solving similar sentiment-related tasks.

Query enhancement, a pivotal methodology in Retrieval-Augmented Generation (RAG) for addressing information scarcity in queries, has garnered increasing research attention. Nevertheless, existing approaches overlook the inherent distinctions between domain-specific knowledge and external factual sources during integration. To bridge this gap, we propose KUG (Knowledge-Update-Generation), a novel RAG framework that leverages internal knowledge semantics to ensure query enhancement efficacy, validates and dynamically updates knowledge representations using external evidence, and achieves systematic integration through knowledge graph embeddings. Extensive experiments on six standard BEIR benchmarks demonstrate that KUG outperforms the state-of-the-art methods, achieving an improvement of 1%-2% in recall metrics. Notably, the framework demonstrates significant performance gains in multi-hop reasoning tasks, advancing the development paradigm for RAG systems. The code will be public soon.

Recent advances in large language models (LLMs) have demonstrated impressive capabilities in code-related tasks such as code generation and automated program repair. Despite their promising performance, most existing approaches for code repair suffer from high training costs or computationally expensive inference. Retrieval-augmented generation (RAG), with its efficient in-context learning paradigm, offers a more scalable alternative. However, conventional retrieval strategies, which are often based on holistic code-text embeddings, fail to capture the structural intricacies of code, resulting in suboptimal retrieval quality. To address the above limitations, we propose ReCode, a fine-grained retrieval-augmented in-context learning framework designed for accurate and efficient code repair. Specifically, ReCode introduces two key innovations: (1) an algorithm-aware retrieval strategy that narrows the search space using preliminary algorithm type predictions; and (2) a modular dual-encoder architecture that separately processes code and textual inputs, enabling fine-grained semantic matching between input and retrieved contexts. Furthermore, we propose RACodeBench, a new benchmark constructed from real-world user-submitted buggy code, which addresses the limitations of synthetic benchmarks and supports realistic evaluation. Experimental results on RACodeBench and competitive programming datasets demonstrate that ReCode achieves higher repair accuracy with significantly reduced inference cost, highlighting its practical value for real-world code repair scenarios.

Recently, Multi-modal Entity Linking (MEL) has attracted increasing attention in the research community due to its significance in numerous multi-modal applications. Video, as a popular means of information transmission, has become prevalent in people’s daily lives. However, most existing MEL methods primarily focus on linking textual and visual mentions or offline videos’ mentions to entities in multi-modal knowledge bases, with limited efforts devoted to linking mentions within online video content. In this paper, we propose a task called Online Video Entity Linking (OVEL), aiming to establish connections between mentions in online videos and a knowledge base with high accuracy and timeliness. To facilitate the research works of (OVEL), we specifically concentrate on live delivery scenarios and construct a live delivery entity linking dataset called (LIVE). Besides, we propose an evaluation metric that considers robustness, timelessness, and accuracy. Furthermore, to effectively handle (OVEL) task, we leverage a memory block managed by a Large Language Model and retrieve entity candidates from the knowledge base to augment LLM performance on memory management. The experimental results prove the effectiveness and efficiency of our method.

Accurate interpretation and visualization of human instructions are crucial for text-to-image (T2I) synthesis. However, current models struggle to capture semantic variations from word order changes, and existing evaluations, relying on indirect metrics like text-image similarity, fail to reliably assess these challenges. This often obscures poor performance on complex or uncommon linguistic patterns by the focus on frequent word combinations. To address these deficiencies, we propose a novel metric called SemVarEffect and a benchmark named SemVarBench, designed to evaluate the causality between semantic variations in inputs and outputs in T2I synthesis. Semantic variations are achieved through two types of linguistic permutations, while avoiding easily predictable literal variations. Experiments reveal that the CogView-3-Plus and Ideogram 2 performed the best, achieving a score of 0.2/1. Semantic variations in object relations are less understood than attributes, scoring 0.07/1 compared to 0.17-0.19/1. We found that cross-modal alignment in UNet or Transformers plays a crucial role in handling semantic variations, a factor previously overlooked by a focus on textual encoders. Our work establishes an effective evaluation framework that advances the T2I synthesis community's exploration of human instruction understanding. Our benchmark and code are available at https://github.com/zhuxiangru/SemVarBench.

Trajectory similarity computation is a fundamental functionality in trajectory data mining, with wide-ranging applications in location-based services. Position embedding, which transforms GPS points into embedding vectors, plays a critical role in learning-based trajectory similarity models. The quality of these embeddings significantly impacts the performance of the models on downstream tasks. Existing methods fail to satisfy all good properties, i.e., global, continuous, unique, and dynamic, thereby limiting the development of trajectory similarity computation in both local and global scenarios. Inspired by linear counting systems, such as the decimal system, we first propose the łambda-base circular system to embed positions on the circle, then introduce the multi-base global embedding method GPE to encode global positions into vectors. Experiments conducted on five real-world datasets with nine baseline methods demonstrate that the GPE achieves state-of-the-art performance across four key evaluations in downstream tasks.

In medical research, clinical trials are pivotal. While prospective clinical research provides a systematic approach to collecting patient data, it grapples with challenges like long durations, increased costs, and most crucially, data scarcity. To address above-mentioned challenge, this paper introduces a novel approach: using cross-table generation to create relevant data. Unlike existing work focused on single-table operations, our method leverages data from multiple sources across various tables, integrating diverse data types and ensuring data consistency across multiple tables. We develop a new framework, MedTransTab, tailored for cross-table tabular data generation in the medical context. This framework extends our previous efforts and is built upon the newly constructed PMC-Struct, derived from an unstructured PMC-patient dataset. Our MedTransTab can generate high-quality patient records, synthesizing detailed biomedical information to align with real or simulated tables from multiple sources. The experiments show that the proposed method significantly improves performance in cross-table tasks. On the PMC-Struct-Plus dataset, we observe an average improvement of 28.85% in data generation and prediction. Similarly, on the Out-Of-Domain (OOD) dataset, there's an average improvement of 22.56%, indicating substantial progress in medical data analysis.

Temporal Knowledge Graph Completion (TKGC) involves predicting and filling in missing facts within time series data, a crucial task with wide-ranging applications across various domains. The dynamic evolution of Temporal Knowledge Graphs (TKGs) adds complexity to this task, making it inherently challenging. Existing research predominantly relies on historical data to complete the missing facts. However, these approaches often overlook the potential of future information and the significance of node weights.To address these challenges, we propose Neo-TKGC, a novel temporal knowledge graph completion model that integrates a graph structure encoding module and a temporal encoding module. The graph structure encoding module introduces node weights to enhance the capabilities of graph neural networks (GNNs) for entity and relation representation learning, implemented using CompGCN. This module can be easily extended to any GNN models utilizing node and edge aggregation. The temporal encoding module leverages both future and historical information to capture relevant contexts and temporal dependencies among entities and relations.By combining node weights and future information, Neo-TKGC achieves more accurate entity and relation representations, thereby improving the model's ability to infer unknown entities. Extensive experiments on three real-world TKGC datasets demonstrate the superior performance of our model compared to existing approaches, achieving at least a 1.7% relative improvement in Hits@1 across most metrics.

Robotic grasping is a fundamental ability for a robot to interact with the environment. Current methods focus on how to obtain a stable and reliable grasping pose in object level, while little work has been studied on part (shape)-wise grasping which is related to fine-grained grasping and robotic affordance. Parts can be seen as atomic elements to compose an object, which contains rich semantic knowledge and a strong correlation with affordance. However, lacking a large part-wise 3D robotic dataset limits the development of part representation learning and downstream applications. In this paper, we propose a new large Language-guided SHape grAsPing datasEt (named LangSHAPE) to promote 3D part-level affordance and grasping ability learning. From the perspective of robotic cognition, we design a two-stage fine-grained robotic grasping framework (named LangPartGPD), including a novel 3D part language grounding model and a partaware grasp pose detection model, in which explicit language input from human or large language models (LLMs) could guide a robot to generate part-level 6-DoF grasping pose with textual explanation. Our method combines the advantages of humanrobot collaboration and LLMs’ planning ability using explicit language as a symbolic intermediate. To evaluate the effectiveness of our proposed method, we perform 3D part grounding and fine-grained grasp detection experiments on both simulation and physical robot settings, following language instructions across different degrees of textual complexity. Results show our method achieves competitive performance in 3D geometry fine-grained grounding, object affordance inference, and 3D part-aware grasping tasks. Our dataset and code are available on our project website https://sites.google.com/view/lang-shape.

Humor is a crucial part of human communication. Understanding humor and generating humorous responses in dialogue can provide natural and empathic human-computer interactions. However, most existing pre-trained language models (PLMs) perform unsatisfactorily in humor generation. On one hand, the serious shortage of humor corpus and datasets pose challenges for constructing models that can understand and generate humorous expressions. On the other hand, humor generation relies on rich knowledge and commonsense, which is often tacit and unspoken. In this paper, we construct the largest Chinese Explainable Humor Response Dataset to date with chain-of-humor and humor mind map annotations, which can be used to comprehensively evaluate as well as improve the humorous response ability of PLMs. We further design humor-related auxiliary tasks to further enhance PLMs' humorous response performance. Extensive evaluations demonstrate that our proposed dataset and auxiliary tasks effectively help PLMs to generate humorous responses, laying the groundwork for future humor research.

Knowledge-grounded dialogue (KGD) learns to generate an informative response based on a given dialogue context and external knowledge (e.g., knowledge graphs; KGs). Recently, the emergence of large language models (LLMs) and pre-training techniques has brought great success to knowledge-grounded dialogue. However, when building KGD systems in real applications, there are various real-world noises that are inevitable to face. For example, the dialogue context might involve perturbations such as misspellings and abbreviations. In addition, KGs typically suffer from incompletion and also might contain erroneous and outdated facts. Such real-world noises pose a challenge to the robustness of KGD systems and hinder their applications in the real world. In this paper, we propose an entity-based contrastive learning framework for improving the robustness of KGD. Specifically, we make use of the entity information in a KGD sample to create both its positive and negative samples which involve semantic-irrelevant and semantic-relevant perturbations, respectively. The contrastive learning framework ensures the KGD model is aware of these two types of perturbations, thus could generate informative responses with the potentially noisy inputs in real applications. Experimental results on three widely-used benchmark datasets show that our method achieves new state-of-the-art performance in terms of automatic evaluation scores, verifying its effectiveness and potentiality. Furthermore, we show that our method is able to generate better responses than comparison models in both the noisy and the few-shot settings.

Named entity recognition aims to identify meaningful words or phrases from text. Traditional NER methods depend heavily on training datasets to learn entity features, which often leads to poor performance in recognizing low-resource entities. Large language models offer a promising solution in low-resource scenarios, but they struggle with entity boundary recognition and often suffer from hallucination issues due to their reliance on general-purpose corpora. In this paper, we propose a novel approach that enhances LLMs’ performance in NER by integrating a boundary regularity extraction algorithm and an explanatory path extraction method. The boundary regularity algorithm leverages ChatGPT’s summarization capabilities to articulate entity boundaries in natural language, improving boundary detection accuracy. The explanatory path method constructs logical reasoning steps from text to entity recognition, guiding LLMs to generate more accurate predictions and mitigating hallucination problems. Extensive experiments on three public datasets demonstrate that our method outperforms current mainstream LLMs in accurately recognizing entities from unstructured text. These findings validate the effectiveness of our approach and its potential to address key challenges in NER.

With the widespread use of GPS devices and the advancement of location-based services, a vast amount of trajectory data has been collected and mined for various applications. Trajectory clustering, which categorizes trajectories into distinct groups, is the fundamental functionality of trajectory data mining. The challenge is how to cluster on a mass of trajectory data efficiently and universally with satisfying results. The raw trajectory clustering algorithms are universal, but trapped in the dilemma between efficiency and desirable results. Other approaches, such as density-based, road network-based, and deep learning-based algorithms, encounter issues like high time complexity, loss of trajectory integrity, reliance on road networks, and data quality during training. To tackle these challenges, we first propose the efficient KMCT (k-Means Clustering of Trajectories) algorithm based on a semantic interpolation transformation to cluster raw trajectories and achieve satisfying results. Additionally, we introduce the DA-KMCT (Density Accelerated k-Means Clustering of Trajectories) algorithm to further boost the clustering process based on trajectory densities and an optimized centroid selecting strategy. Moreover, we present a novel clustering evaluation method called IOD, which efficiently estimates clustering results on large-scale datasets with linear time complexity. Experimental results on real-world datasets demonstrate that KMCT and DA-KMCT outperform five related methods in terms of clustering quality and time efficiency, and the proposed IOD evaluation shows a strong correlation with the Silhouette Coefficient, offering a reliable and efficient alternative for evaluating clustering results.

Pre-trained language models (PLMs) have established the new paradigm in the field of NLP. For more powerful PLMs, one of the most popular and successful ways is to continuously scale up sizes of the models and the pre-training corpora. These large corpora, typically obtained by converging smaller ones from multiple sources, are thus growing increasingly diverse. However, colossal converged corpora don't always enhance PLMs' performance. In this paper, we identify the disadvantage of heterogeneous corpora from multiple sources for pre-training PLMs. Towards coordinated pre-training on diverse corpora, we further propose Source Prompt (SP), which explicitly prompt the model with the source of data at the pre-training and fine-tuning stages. Extensive experimental results show that pre-training PLMs with SP on diverse corpora significantly improves performance in various downstream tasks.

Multimodal knowledge bases (MMKBs) provide cross-modal aligned knowledge crucial for multimodal tasks. However, the images in existing MMKBs are generally collected for entities in encyclopedia knowledge graphs. Therefore, detailed groundings of visual semantics with linguistic concepts are lacking, which are essential for the visual concept cognition ability of multimodal models. Addressing this gap, we introduce M2 ConceptBase, the first concept-centric MMKB. M2 ConceptBase models concepts as nodes with associated images and detailed textual descriptions. We propose a context-aware multimodal symbol grounding approach to align concept-image and concept-description pairs using context information from image-text datasets. Comprising 951K images and 152K concepts, M2 ConceptBase links each concept to an average of 6.27 images and a single description, ensuring comprehensive visual and textual semantics. Human studies confirm more than 95% alignment accuracy, underscoring its quality. Additionally, our experiments demonstrate that M2 ConceptBase significantly enhances VQA model performance on the OK-VQA task. M2 ConceptBase also substantially improves the fine-grained concept understanding capabilities of multimodal large language models through retrieval augmentation in two concept-related tasks, highlighting its value.

Visual language navigation (VLN) is one of the important research in embodied AI. It aims to enable an agent to understand the surrounding environment and complete navigation tasks. VLN instructions could be categorized into coarse-grained and fine-grained commands. Fine-grained command describes a whole task with subtasks step-by-step. In contrast, coarse-grained command gives an abstract task description, which more suites human habits. Most existing work focuses on the former kind of instruction in VLN tasks, ignoring the latter abstract instructions belonging to daily life scenarios. To overcome the above challenge in abstract instruction, we attempt to consider coarse-grained instruction in VLN by event knowledge enhancement. Specifically, we first propose a prompt-based framework to extract an event knowledge graph (named VLN-EventKG =) for VLN integrally over multiple mainstream benchmark datasets. Through small and large language model collaboration, we realize knowledge-enhanced navigation planning (named EventNav) for VLN tasks with coarse-grained instruction input. Additionally, we design a novel dynamic history backtracking module to correct potential error action planning in real time. Experimental results in various public benchmarks show our knowledge-enhanced method has superiority in coarse-grained-instruction VLN using our proposed VLN-EventKG with over 5% improvement in success rate. Our project is available at https://sites.google.com/view/vln-eventkg

Biomedical Named Entity Recognition (BioNER) plays a crucial role in automatically identifying specific categories of entities from biomedical texts. Currently, region-based methods have shown promising performance in BioNER. However, existing paradigms in the region-based methods suffer from inherent limitations, including the generation of negative samples, and the ignorance of token dependencies. To overcome these limitations, we propose a new paradigm, implemented as Token Cascade Tagger (TCT), which combines span identification and category classification. The TCT utilizes category information to enhance the correlation between the heads and tails of entities, effectively reducing the generation of negative samples. Additionally, we introduce a Token Dependency Tagger (TDT) that captures token dependencies within entity spans by identifying the longest span in a sentence. The TDT filters out incorrect spans and further improves the accuracy of span detection obtained from the TCT. Furthermore, we employ a multi-task learning framework to optimize both the TCT and TDT, leading to superior performance in BioNER. Extensive experiments on publicly available biomedical datasets demonstrate our method outperforms the previous state-of-the-art methods, achieving 92.44%, 92.54%, and 81.26% on NCBI-Disease, BC5CDR, and GENIA, respectively, in terms of F1 score.

Web scraping is a powerful technique that extracts data from websites, enabling automated data collection, enhancing data analysis capabilities, and minimizing manual data entry efforts.Existing methods, wrappers-based methods suffer from limited adaptability and scalability when faced with a new website, while language agents, empowered by large language models (LLMs), exhibit poor reusability in diverse web environments.In this work, we introduce the paradigm of generating web scrapers with LLMs and propose AUTOSCRAPER, a two-stage framework that can handle diverse and changing web environments more efficiently.AUTOSCRAPER leverages the hierarchical structure of HTML and similarity across different web pages for generating web scrapers.Besides, we propose a new executability metric for better measuring the performance of web scraper generation tasks.We conduct comprehensive experiments with multiple LLMs and demonstrate the effectiveness of our framework.Our work is now open-source.

Haiquan Zhao, Lingyu Li, Shisong Chen, Shuqi Kong, Jiaan Wang, Kexin Huang, Tianle Gu, Yixu Wang, Jian Wang, Liang Dandan, Zhixu Li, Yan Teng, Yanghua Xiao, Yingchun Wang. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Large language models(LLMs) have achieved remarkable performance on various tasks. However, LLMs suffer from severe limitations in domain generalisation, primarily due to inherent limitations. Closed-source LLMs face constraints in fine-tuning, while open-source LLMs contend with the scarcity of domain-specific data. Moreover, LLMs often prioritize addressing standard patterns, inadvertently neglecting intricate and domain-specific patterns. This preference hampers effective domain generalization. In this paper, inspired by curriculum learning, we explore a cascade instruction tuning method to train a domain-specific LLMs that can excel in a broad application such as information extraction. Taking biomedical name entity recognition(BioNER) as a case study, we show how to cultivate general LLMs into domain-specific LLMs with limited domain data and address the complex pattern for BioNER. To validate our method, we construct NER INSTRUCTIONS, the largest and broadest benchmark sourced from 55 publicly available NER datasets across 17 domains. We conduct extensive experiments on the dataset, and the results demonstrate the effectiveness of our proposed framework in downstream task generalisation and its ability to tackle intricate patterns.

Extracting entities and relations from text is a significant task of information extraction. Existing extraction models often straightforwardly produce their confident prediction results without any reconsideration or double-checking, resulting in avoidable mistakes and sub-optimal performance. In this paper, we propose a novel coarse-to-fine extraction framework, which first extracts high-potential relations as well as entities via knowledge distillation, and then rechecks the predictions via handcrafted natural language inference (NLI) task in a fine-grained manner. Specifically, based on the knowledge distillation mechanism, we train multiple teacher models iteratively through an adaptive loss function for making one teacher concentrate more on the data that others are incompetent for. Then, these complementary teacher models are utilized to provide valuable soft-label information for training a considerate student model, enabling it to generate reliable preliminary predictions. Further, these generated potential relations and entities are formulated as hypotheses, together with the original sentences as premises, serving as the input for an NLI model. Considering the linguistic diversity of relational expression, we automatically generate various semantic templates for hypotheses through an$\mathcal{N}$-gram mining strategy. Moreover, due to the existence of multi-fact sentences, a relation-guided Gaussian attention is designed to reduce the gap between the single-relation hypothesis and the multi-relation premise. To implement efficient training, we also develop several ways to generate high-quality negative samples, which help the NLI model learn to identify errors. Experimental results show that the proposed method is effective and outperforms other strong baselines on public benchmarks.

Humor, deeply rooted in societal meanings and cultural details, poses a unique challenge for machines. While advances have been made in natural language processing, real-world humor often thrives in a multi-modal context, encapsulated distinctively by memes. This paper poses a particular emphasis on the impact of multi-images on meme captioning. After that, we introduce the XMeCap framework, a novel approach that adopts supervised fine-tuning and reinforcement learning based on an innovative reward model, which factors in both global and local similarities between visuals and text. Our results, benchmarked against contemporary models, manifest a marked improvement in caption generation for both single-image and multi-image memes, as well as different meme categories. XMeCap achieves an average evaluation score of 75.85 for single-image memes and 66.32 for multi-image memes, outperforming the best baseline by 3.71% and 4.82%, respectively. This research not only establishes a new frontier in meme-related studies but also underscores the potential of machines in understanding and generating humor in a multi-modal setting.

Medical dialogue models predominantly emphasize generating coherent and clinically accurate responses. However, in many clinical scenarios, time plays a pivotal role, often dictating subsequent patient management and interventions. Recognizing the latent importance of temporal dynamics, this paper introduces a novel dimension to medical dialogues: timestamps. We advocate that the integration of time-sensitive directives can profoundly impact medical advice, using an illustrative example of post-surgery care with and without timestamps. Our contributions are three-fold: Firstly, we highlight the intrinsic significance of timestamps in medical conversations, marking a paradigm shift in dialogue modeling. Secondly, we present an innovative dataset and framework explicitly tailored for time-stamped medical dialogues, facilitating the model to not only provide medical counsel but also chronologically outline care regimens. Lastly, empirical evaluations indicate our method's proficiency in time-stamped tasks and reveal an uptick in performance in broader medical Q&A domains. Through our endeavors, we aspire to set new benchmarks in patient-centric and time-sensitive medical dialogue systems.

Question understanding is an important issue to the success of a Knowledge-based Question Answering (KBQA) system.However, the existing study does not pay enough attention to this issue given that the questions in the existing KBQA datasets are usually expressed in simple and straightforward way. This is not in line with the actual linguistic conventions, which often use a lot of modifiers. To facilitate the study on evaluating and enhancing the question understanding ability of the KBQA systems, this paper proposes to construct a complex-modified question-answering (XMQAs) dataset based on existing KBQA datasets. With the help of knowledge bases and dictionaries, three kinds of modifiers are defined and applied to original simple-expressed questions. These modifiers could make the expression of these questions complex without changing their semantics. Based on XMQAs, we then propose a novel question understanding algorithm upon existing KBQA models, which greatly improves the robustness of their question understanding abilities. We conduct extensive experiments on XMQAs and two widely acknowledged KBQA datasets. The empirical results demonstrate that our proposed algorithm can improve the performance of KBQA models on not only the complex-modified questions, but also simple-expressed questions.

Embodied AI is one of the most popular studies in artificial intelligence and robotics, which can effectively improve the intelligence of real-world agents (i.e. robots) serving human beings. Scene knowledge is important for an agent to understand the surroundings and make correct decisions in the varied open world. Currently, knowledge base for embodied tasks is missing and most existing work use general knowledge base or pre-trained models to enhance the intelligence of an agent. For conventional knowledge base, it is sparse, insufficient in capacity and cost in data collection. For pre-trained models, they face the uncertainty of knowledge and hard maintenance. To overcome the challenges of scene knowledge, we propose a scene-driven multimodal knowledge graph (Scene-MMKG) construction method combining conventional knowledge engineering and large language models. A unified scene knowledge injection framework is introduced for knowledge representation. To evaluate the advantages of our proposed method, we instantiate Scene-MMKG considering typical indoor robotic functionalities (Manipulation andMobility), namedManipMob-MMKG. Comparisons in characteristics indicate our instantiated ManipMob-MMKG has broad superiority on data-collection efficiency and knowledge quality. Experimental results on typical embodied tasks show that knowledge-enhanced methods using our instantiated ManipMob-MMKG can improve the performance obviously without re-designing model structures complexly.

Document-level Relation Extraction (RE) is a promising task aiming at identifying relations of multiple entity pairs in a document. Compared with the sentence-level counterpart, it has raised two significant challenges: a) In most cases, a relational fact can be adequately expressed via a small subset of sentences from the document, namely evidence. But the traditional method cannot model such strong semantic correlations between evidence sentences that collaborate to describe a specific relation; b) The data of this task is extremely long-tail in terms of too many NA instances and imbalanced relational types. Such data can mislead the tail prediction bias to the head categories in the RE model. In this paper, we present a novelEvidence reasoning andCurriculum learning method forDocRE(DRE-EC) to address these challenges. Particularly, we first formulate evidence extraction as a sequential decision problem through a crafted reinforcement learning mechanism with an efficient path searching strategy to reduce the action space. Providing the evidence for each entity pair as a customized-filtered document in advance helps infer the relations better. To address the long-tail issue, we further develop a hybrid curriculum learning method at the NA-level (NC) and relation-level (RC) with our customized difficulty measure score. In NC, the NA samples are scheduled in an easy-to-hard scheme and gradually added, resulting in the data distribution from ideal and balanced to real and unbalanced. In RC, the scheme is switched into hard-to-easy to enhance the hard and tail samples. In addition, we propose a new Equalization adaptive Focal Loss(EFLoss) that can adjust to the changing data distribution and focus more on the tail categories. We conduct various experiments on two document-level RE benchmarks and achieve a remarkable improvement over previous competitive baselines. Furthermore, we provide detailed analyses of the advantages and effectiveness of our method.

Recent years have witnessed the resurgence of knowledge engineering which is featured by the fast growth of knowledge graphs. However, most of existing knowledge graphs are represented with pure symbols, which hurts the machine's capability to understand the real world. The multi-modalization of knowledge graphs is an inevitable key step towards the realization of human-level machine intelligence. The results of this endeavor are Multi-modal Knowledge Graphs (MMKGs). In this survey on MMKGs constructed by texts and images, we first give definitions of MMKGs, followed with the preliminaries on multi-modal tasks and techniques. We then systematically review the challenges, progresses and opportunities on the construction and application of MMKGs respectively, with detailed analyses of the strengths and weaknesses of different solutions. We finalize this survey with open research problems relevant to MMKGs.

Taxonomy is formulated as directed acyclic graphs or trees of concepts that support many downstream tasks. Many new coming concepts need to be added to an existing taxonomy. The traditional taxonomy expansion task aims only at finding the best position for new coming concepts in the existing taxonomy. However, they have two drawbacks when being applied to the real-scenarios. The previous methods suffer from low-efficiency since they waste much time when most of the new coming concepts are indeed noisy concepts. They also suffer from low-effectiveness since they collect training samples only from the existing taxonomy, which limits the ability of the model to mine more hypernym-hyponym relationships among real concepts. This paper proposes a pluggable framework called Generative Adversarial Network for Taxonomy Entering Evaluation (GANTEE) to alleviate these drawbacks. A generative adversarial network is designed in this framework by discriminative models to alleviate the first drawback and the generative model to alleviate the second drawback. Two discriminators are used in GANTEE to provide long-term and short-term rewards, respectively. Moreover, to further improve the efficiency, pre-trained language models are used to retrieve the representation of the concepts quickly. The experiments on three real-world large-scale datasets with two different languages show that GANTEE improves the performance of the existing taxonomy expansion methods in both effectiveness and efficiency.

Semi-Supervised Relation Extraction aims at learning well-performed RE models with limited labeled and large-scale unlabeled data. Existing methods mainly suffer from semantic drift and insufficient supervision, which severely limit the performance. To address these problems, recent work tends to design dual modules to work cooperatively for mutual enhancement. However, the consensus of two modules greatly restricts the model from exploring diverse relation expressions in unlabeled set, which hinders the performance as well as model generalization. To tackle this problem, in this paper, we propose a novel competition-based method AdvSRE. We set up a challenging minimax game on unlabeled data between two modules, Generator and Discriminator, and assign them with conflicting objectives. During the competition game, one module may find any possible chance to beat the other, which develops two modules' abilities until relation expressions cannot be further explored. To exploit label information, Discriminator is further asked to predict specific relation for each sentence. Experiment results on two benchmarks show new state-of-the-art performance over baselines, demonstrating the effectiveness of proposed AdvSRE.

Jiaan Wang, Fandong Meng, Duo Zheng, Yunlong Liang, Zhixu Li, Jianfeng Qu, Jie Zhou. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Recent years, Pre-trained Language models (PLMs) have swept into various fields of artificial intelligence and achieved great success. However, most PLMs, such as T5 and GPT3, have a huge amount of parameters, fine-tuning them is often expensive and time consuming, and storing them takes up a lot of space. Therefore, it is necessary to adopt a parameter-efficient approach to reduce parameters of PLMs in fine-tuning without compromising their performance in downstream tasks. In this paper, we design a novel adapter which only acts on self-attention outputs in PLMs. This adapter adopts element-wise linear transformation using Hadamard product, hence named as Hadamard adapter, requires the fewest parameters compared to previous parameter-efficient adapters. In addition, we also summarize some tuning patterns for Hadamard adapter shared by various downstream tasks, expecting to provide some guidance for further parameter reduction with shared adapters in future studies. The experiments conducted on the widely-used GLUE benchmark with several SOTA PLMs prove that the Hadamard adapter achieves competitive performance with only 0.033% parameters compared with full fine-tuning, and it has the fewest parameters compared with other adapters. Moreover, we further find that there is also some redundant layers in the Hadamard adapter which can be removed to achieve more parameter efficiency with only 0.022% parameters.

Large language models (LLMs) have gained widespread adoption in various natural language processing tasks, including question answering and dialogue systems. However, a major drawback of LLMs is the issue of hallucination, where they generate unfaithful or inconsistent content that deviates from the input source, leading to severe consequences. In this paper, we propose a robust discriminator named RelD to effectively detect hallucination in LLMs' generated answers. RelD is trained on the constructed RelQA, a bilingual question-answering dialogue dataset along with answers generated by LLMs and a comprehensive set of metrics. Our experimental results demonstrate that the proposed RelD successfully detects hallucination in the answers generated by diverse LLMs. Moreover, it performs well in distinguishing hallucination in LLMs' generated answers from both in-distribution and out-of-distribution datasets. Additionally, we also conduct a thorough analysis of the types of hallucinations that occur and present valuable insights. This research significantly contributes to the detection of reliable answers generated by LLMs and holds noteworthy implications for mitigating hallucination in the future work.

Multi-modal knowledge graphs (MMKGs) combine different modal data (e.g., text and image) for a comprehensive understanding of entities. Despite the recent progress of large-scale MMKGs, existing MMKGs neglect the multi-aspect nature of entities, limiting the ability to comprehend entities from various perspectives.In this paper, we construct AspectMMKG, the first MMKG with aspect-related images by matching images to different entity aspects. Specifically, we collect aspect-related images from a knowledge base, and further extract aspect-related sentences from the knowledge base as queries to retrieve a large number of aspect-related images via an online image search engine. Finally, AspectMMKG contains 2,380 entities, 18,139 entity aspects, and 645,383 aspect-related images. We demonstrate the usability of AspectMMKG in entity aspect linking (EAL) downstream task and show that previous EAL models achieve a new state-of-the-art performance with the help of AspectMMKG.To facilitate the research on aspect-related MMKG, we further propose an aspect-related image retrieval (AIR) model, that aims to correct and expand aspect-related images in AspectMMKG.We train an AIR model to learn the relationship between entity image and entity aspect-related images by incorporating entity image, aspect, and aspect image information. Experimental results indicate that the AIR model could retrieve suitable images for a given entity w.r.t different aspects.

Nested named entity recognition (Nested NER) aims to identify entities with nested structures from the given text, which is a fundamental task in Natural Language Processing. The region-based approach is the current mainstream approach, which first generates candidate spans and then classifies them into predefined categories. However, this method suffers from several drawbacks, including over-reliance on span representation, vulnerability to unbalanced category distribution, and inaccurate span boundary detection. To address these problems, we propose to model the nested NER problem into a head-tail mapping problem, namely, HTMapper, which detects head boundaries first and then models a conditional mapping from head to tail under a given category. Based on this mapping, we can find corresponding tails under different categories for each detected head by enumerating all entity categories. Our approach directly models the head boundary and tail boundary of entities, avoiding over-reliance on the span representation. Additionally, Our approach utilizes category information as an indicator signal to address the imbalance of category distribution during category prediction. Furthermore, our approach enhances the detection of span boundaries by capturing the correlation between head and tail boundaries. Extensive experiments on three nested NER datasets and two flat NER datasets demonstrate that our HTMapper achieves excellent performance with F1 scores of 89.09%, 88.30%, 81.57% on ACE2004,ACE2005, GENIA, and 94.26%, 91.40% on CoNLL03, OntoNotes, respectively.

Humor understanding is an important and challenging research in natural language processing. As the popularity of pre-trained language models (PLMs), some recent work makes preliminary attempts to adopt PLMs for humor recognition and generation. However, these simple attempts do not substantially answer the question: whether PLMs are capable of humor understanding? This paper is the first work that systematically investigates the humor understanding ability of PLMs. For this purpose, a comprehensive framework with three evaluation steps and four evaluation tasks is designed. We also construct a comprehensive Chinese humor dataset, which can fully meet all the data requirements of the proposed evaluation framework. Our empirical study on the Chinese humor dataset yields some valuable observations, which are of great guiding value for future optimization of PLMs in humor understanding and generation.

Image-text retrieval is a challenging cross-modal task that arouses much attention. While the traditional methods cannot break down the barriers between different modalities, Vision-Language Pre-trained (VLP) models greatly improve image-text retrieval performance based on massive image-text pairs. Nonetheless, the VLP-based methods are still prone to produce retrieval results that cannot be cross-modal aligned with entities. Recent efforts try to fix this problem at the pre-training stage, which is not only expensive but also unpractical due to the unavailable of full datasets. In this paper, we novelly propose a lightweight and practical approach to align cross-modal entities for image-text retrieval upon VLP models only at the fine-tuning and re-ranking stages. We employ external knowledge and tools to construct extra fine-grained image-text pairs, and then emphasize cross-modal entity alignment through contrastive learning and entity-level mask modeling in fine-tuning. Besides, two re-ranking strategies are proposed, including one specially designed for zero-shot scenarios. Extensive experiments with several VLP models on multiple Chinese and English datasets show that our approach achieves state-of-the-art results in nearly all settings.

Cross-Lingual Summarization (CLS) aims at generating summaries in one language for the given documents in another language. CLS has attracted wide research attention due to its practical significance in the multi-lingual world. Though great contributions have been made, existing CLS works typically focus on short documents, such as news and guides. Different from these short texts, long documents such as academic articles usually discuss complicated subjects and consist of thousands of words, making them non-trivial to process and summarize. To promote CLS research on long documents, we construct Perseus, the first long-document CLS dataset which collects about 94K Chinese scientific documents paired with English summaries. The average length of documents in Perseus is more than 2000 tokens. As a preliminary study on long-document CLS, we build and evaluate various CLS baselines, including pipeline and end-to-end methods. Experimental results on Perseus show the superiority of the end-to-end baseline, which performs the best among all methods. Furthermore, to provide a deeper understanding, we manually analyze the model outputs and discuss specific challenges faced by current approaches. We hope that our work could benchmark long-document CLS and benefit future studies.

Simile interpretation is a crucial task in natural language processing.Nowadays, pre-trained language models (PLMs) have achieved stateof-the-art performance on many tasks.However, it remains under-explored whether PLMs can interpret similes or not.In this paper, we investigate the ability of PLMs in simile interpretation by designing a novel task named Simile Property Probing, i.e., to let the PLMs infer the shared properties of similes.We construct our simile property probing datasets from both general textual corpora and humandesigned questions, containing 1,633 examples covering seven main categories.Our empirical study based on the constructed datasets shows that PLMs can infer similes' shared properties while still underperforming humans.To bridge the gap with human performance, we additionally design a knowledge-enhanced training objective by incorporating the simile knowledge into PLMs via knowledge embedding methods.Our method results in a gain of 8.58% in the probing task and 1.37% in the downstream task of sentiment classification.The datasets and code are publicly available at https://github.com/Abbey4799/PLMs- Interpret-Simile.

Xuwu Wang, Junfeng Tian, Min Gui, Zhixu Li, Rui Wang, Ming Yan, Lihan Chen, Yanghua Xiao. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Neural Relation Extraction (RE) models need large amounts of labeled data for effective training, which mainly comes from automatically labeling by Distant Supervision (DS). Though fast and easy, the label shift problem inevitably happens, i.e., the label distribution of DS-generated training set is quite different from that of the real world (i.e. test set). According to our observations, label shift not only leads to performance diminishment, but also hinders the reliability of DS-RE models by causing bad confidence estimation. In this paper, we make contributions by answering the following three questions: 1) How to improve performance of DS-RE models under label shift? 2) How to make sure their reliability under label shift? 3) How to improve both performance and reliability for DS-RE models under label shift? To the best of our knowledge, this is the first paper to study the performance as well as reliability of DS-RE models under label shift. Experiment results show significant improvements on two real-world datasets and six popular neural RE models, making a step further towards high-performance and reliable RE system under real-world label-shift conditions.

In this paper, we study the problem of on-demand food delivery in a new setting where two groups of workers -- riders and taxi drivers (drivers for short) -- cooperate with each other for better service. The riders are responsible for the first and the last mile, and the drivers are in charge of the cross-community transportation. We show this problem is generally NP-hard by a reduction from the well-known 3-dimensional matching (3DM). To tackle with this problem, we first reduce it to the maximum independent set problem and use a simple greedy strategy to design an approximate algorithm which has a polynomial time. Considering the exponents in the polynomial are not very small, we then transform the 3DM into two rounds of 2-dimensional matching and propose a fast algorithm to solve it. Though 3DM problem is NP-hard, we find the cooperation between riders and drivers form a special tripartite graph, based on which we construct a flow network and employ the min-cost max-flow algorithm to efficiently compute the exact solution. We conduct extensive experiments to show the efficiency and the effectiveness of our proposed algorithms.

Document-level Relation Extraction (RE) is a promising task aiming at identifying relations of multiple entity pairs in a document. However, in most cases, a relational fact can be expressed enough via a small subset of sentences from the document, namely evidence sentence. Moreover, there often exist strong semantic correlations between evidence sentences that collaborate together to describe a specific relation. To address these challenges, we propose a novel evidence-aware model for document-level RE. Particularly, we formulate evidence sentence selection as a sequential decision problem through a crafted reinforcement learning mechanism. Considering the explosive search space of our agent, an efficient path searching strategy is executed on the converted document graph to heuristically obtain hopeful sentences and feed them to reinforcement learning. Finally, each entity pair owns a customized-filtered document for further inferring the relation between them. We conduct various experiments on two document-level RE benchmarks and achieve a remarkable improvement over previous competitive baselines, verifying the effectiveness of our method.

We present CLIDSUM, a benchmark dataset towards building cross-lingual summarization systems on dialogue documents.It consists of 67k+ dialogue documents and 112k+ annotated summaries in different target languages.Based on the proposed CLIDSUM, we introduce two benchmark settings for supervised and semi-supervised scenarios, respectively.We then build various baseline systems in different paradigms (pipeline and end-to-end) and conduct extensive experiments on CLIDSUM to provide deeper analyses.Furthermore, we propose mDIALBART which extends mBART via further pre-training, where the multiple objectives help the pre-trained model capture the structural characteristics as well as key content in dialogues and the transformation from source to the target language.Experimental results show the superiority of mDIALBART, as an end-to-end model, outperforms strong pipeline models on CLIDSUM.Finally, we discuss specific challenges that current approaches faced with this task and give multiple promising directions for future research.

Entity typing aims to assign types to the entity mentions in given texts.The traditional classification-based entity typing paradigm has two unignorable drawbacks: 1) it fails to assign an entity to the types beyond the predefined type set, and 2) it can hardly handle few-shot and zero-shot situations where many long-tail types only have few or even no training instances.To overcome these drawbacks, we propose a novel generative entity typing (GET) paradigm: given a text with an entity mention, the multiple types for the role that the entity plays in the text are generated with a pre-trained language model (PLM).However, PLMs tend to generate coarse-grained types after finetuning upon the entity typing dataset.In addition, only the heterogeneous training data consisting of a small portion of human-annotated data and a large portion of auto-generated but low-quality data are provided for model training.To tackle these problems, we employ curriculum learning (CL) to train our GET model on heterogeneous data, where the curriculum could be self-adjusted with the self-paced learning according to its comprehension of the type granularity and data heterogeneity.Our extensive experiments upon the datasets of different languages and downstream tasks justify the superiority of our GET model over the state-ofthe-art entity typing models.The code has been released on https://github.com/siyuyuan/GET.

Multimodal named entity recognition (MNER) aims to detect and classify named entities in multimodal scenarios. It requires bridging the gap between natural language and visual context, which presents two-fold challenges: the cross-modal alignment is diversified, and the cross-modal interaction is sometimes implicit. Existing MNER methods are vulnerable to some implicit interactions and are prone to overlook the involved significant features. To tackle this problem, we novelly propose to refine the cross-modal attention by identifying and highlighting some task-salient features. The saliency of each feature is measured according to its correlation with the expanded entity label words derived from external knowledge bases. We further propose an end-to-end Transformer-based MNER framework, which holds neater architecture yet achieves better performance than previous methods. Extensive experiments are conducted to validate the merits of our method. Moreover, our method reveals a significant advantage in data efficiency and generalization ability.

Sports game summarization aims at generating sports news from live commentaries. However, existing datasets are all constructed through automated collection and cleaning processes, resulting in a lot of noise. Besides, current works neglect the knowledge gap between live commentaries and sports news, which limits the performance of sports game summarization. In this paper, we introduce K-SportsSum, a new dataset with two characteristics: (1) K-SportsSum collects a large amount of data from massive games. It has 7,854 commentary-news pairs. To improve the quality, K-SportsSum employs a manual cleaning process; (2) Different from existing datasets, to narrow the knowledge gap, K-SportsSum further provides a large-scale knowledge corpus that contains the information of 523 sports teams and 14,724 sports players. Additionally, we also introduce a knowledge-enhanced summarizer that utilizes both live commentaries and the knowledge to generate sports news. Extensive experiments on K-SportsSum and SportsSum datasets show that our model achieves new state-of-the-art performances. Qualitative analysis and human study further verify that our model generates more informative sports news.

Abstract Cross-lingual summarization is the task of generating a summary in one language (e.g., English) for the given document(s) in a different language (e.g., Chinese). Under the globalization background, this task has attracted increasing attention of the computational linguistics community. Nevertheless, there still remains a lack of comprehensive review for this task. Therefore, we present the first systematic critical review on the datasets, approaches, and challenges in this field. Specifically, we carefully organize existing datasets and approaches according to different construction methods and solution paradigms, respectively. For each type of dataset or approach, we thoroughly introduce and summarize previous efforts and further compare them with each other to provide deeper analyses. In the end, we also discuss promising directions and offer our thoughts to facilitate future research. This survey is for both beginners and experts in cross-lingual summarization, and we hope it will serve as a starting point as well as a source of new ideas for researchers and engineers interested in this area.

Massive trajectory data stem from the prevalence of equipment-supporting GPS and wireless communication technology. Especially, activity trajectory from Location-based Social Network (LBSN) endows traditional trajectory data with additional user semantic activities, e.g., visiting work/home/entertainment places. Measuring the similarity between activity trajectories is to compare their proximity in multiple dimensions such as time, location, and semantics. In this way, we can mine implicit user preference and apply it to route planning, POI recommendation or any other online tasks. The key challenge of comparing activity trajectories (i.e., computing their similarity) lies in two aspects. One is the uneven sampling rate in both time and space. The other is the discrepancy of individual activities. Previous effort alleviates the issue of uneven sampling rate via trajectory complements, which is limited to spatial-temporal information. In this paper, we propose to learn a representation for one activity trajectory by jointly considering the spatio-temporal characteristics and the activity semantics. The similarity of two trajectories is computed by weighting individual trajectory points and contextual features with multi-level attention mechanisms. In specific, we propose a point-level and feature-level attention mechanism to adaptively select critical elements and contextual factors for learning trajectory representation. Our proposed approach, called At2vec, demonstrates better performance than existing baselines in extensive experimental evaluation on real trajectory databases.

It has been an important task for recommender systems to suggest satisfying activities to a group of users in peoples daily social life. The major challenge in this task is how to aggregate personal preferences of group members to infer the decision of a group. In this paper, we propose a novel end-to-end group recommender system named CAGR (short for Centrality-Aware Group Recommender), which takes the Bipartite Graph Embedding Model (BGEM), the self-attention mechanism and Graph Convolutional Networks (GCNs) as basic building blocks to learn group and user representations in a unified way. Specifically, we first extend BGEM to model group-item interactions, and then in order to overcome the sparsity of the interaction data generated by occasional groups, we propose a self-attentive mechanism to represent groups based on the group members. To further alleviate the group data sparsity problem, we propose two model optimization approaches to exploit an and integrate the user-item interaction data. To overcome the sparsity issue of user-item interaction data, we extend GCNs to leverage the social network to enhance user representation learning. We create two large-scale benchmark datasets and conduct extensive experiments on them. The experimental results show the superiority of our proposed CAGR.

Next Point-of-Interest (POI) recommendation which is of great value to both users and POI holders is a challenging task since complex sequential patterns and rich contexts are contained in extremely sparse user check-in data. Recently proposed embedding techniques have shown promising results in alleviating the data sparsity issue by modeling context information, and Recurrent Neural Network (RNN) has been proved effective in the sequential prediction. However, existing next POI recommendation approaches train the embedding and network model separately, which cannot fully leverage rich contexts. In this paper, we propose a novel unified neural network framework, named NeuNext, which leverages POI context prediction to assist next POI recommendation by joint learning. Specifically, the Spatio-Temporal Gated Network (STGN) is proposed to model personalized sequential patterns for users’ long and short term preferences in the next POI recommendation. In the POI context prediction, rich contexts on POI sides are used to construct graph, and enforce the smoothness among neighboring POIs. Finally, we jointly train the POI context prediction and the next POI recommendation to fully leverage labeled and unlabeled data. Extensive experiments on real-world datasets show that our method outperforms other approaches for next POI recommendation in terms of Accuracy and MAP.

Sports game summarization aims to generate news articles from live text commentaries. A recent state-of-the-art work, SportsSum, not only constructs a large benchmark dataset, but also proposes a two-step framework. Despite its great contributions, the work has three main drawbacks: 1) the noise existed in SportsSum dataset degrades the summarization performance; 2) the neglect of lexical overlap between news and commentaries results in low-quality pseudo-labeling algorithm; 3) the usage of directly concatenating rewritten sentences to form news limits its practicability. In this paper, we publish a new benchmark dataset SportsSum2.0, together with a modified summarization framework. In particular, to obtain a clean dataset, we employ crowd workers to manually clean the original dataset. Moreover, the degree of lexical overlap is incorporated into the generation of pseudo labels. Further, we introduce a reranker-enhanced summarizer to take into account the fluency and expressiveness of the summarized news. Extensive experiments show that our model outperforms the state-of-the-art baseline.

Zero pronoun resolution aims at recognizing dropped pronouns and pointing out their anaphoric mentions, while non-zero coreference resolution targets at clustering mentions referring to the same entity.Existing efforts often deal with the two problems separately regardless of their close essential correlations.In this paper, we investigate the possibility of jointly solving zero pronoun resolution and coreference resolution via a novel end-to-end neural model.Specifically, we design a gapmasked self-attention model that encodes gaps and tokens in the same space, where gaps could capture valuable contextual information according to their surrounding tokens while tokens could maintain original sequential information without disturbance.Additionally, we also propose a two-stage interaction mechanism to make full use of the exclusive relationship between zero pronouns and mentions.Our empirical study conducted on the OntoNotes 5.0 Chinese dataset shows that our model could outperform corresponding stateof-the-art approaches on both tasks.

Knowledge graph embedding (KGE) plays an important role in downstream tasks, such as question answering, recommendation system, and entity recognition. Most existing KGE methods focus on modeling static knowledge graphs. However, many knowledge graphs are incremental in reality. Existing KGE methods are time-consuming to update the embedding space incrementally, and have difficulty in keeping the timeliness of the knowledge graph embedding. To address this problem, we propose a novel knowledge graph embedding method by rotating on hyperplane (RotatH), which supports updating the embedding space incrementally and ensures the timeliness and accuracy of knowledge graph embedding. Specifically, our proposed method first employs relation-specific hyperplanes to update the incremental entities into the trained vector space efficiently. Meanwhile, by combining hyperplane and rotation, our method can deal with complex relations, such as many-to-many and symmetry relations, and has high performance in both incremental and static environments. Moreover, our method introduces a mean-based method to constraint the density of incremental entities. We conduct extensive link prediction experiments on two real-world incremental datasets and two benchmark datasets. The experimental results show that our model incrementally updates embedding space efficiently and outperforms static models on benchmarks.

Data incompleteness is a common data quality problem in databases. Recent work proposes to retrieve missing string values from the World Wide Web for higher imputation recall, but on the other hand, takes the risk of introducing web noises into the imputation results. So far there lacks an effective way to control the quality of web-based data imputation, given the complexity of the quality model and lacking of enough ground truth data. In this article, an EM-based quality model is first built for web-based data imputation which investigates three key factors jointly, i.e., precision of web sources, correlation among web sources, and precision and recall of the employed extractors. However, the accuracy of the EM-based quality model could be harmed when the EM (Expectation Maximization) assumption that “the majority agree on the truth” does not hold in some cases. To solve this problem, we introduce crowd intervention to help improve the quality model. While a straightforward but expensive way is to let the crowd to identify all these undesirable cases and provide the right imputation values for these blanks, a most crowd-economic way is to select a small set of blanks for crowd-based imputation, whose results could help to adjust the EM-based quality model towards a better one. To achieve this, an adaptive blank selection strategy is proposed to select a sequence of blanks for crowd-based imputation. Also, we work on finding a proper time to stop further crowd intervention for the balance of crowd efficiency and quality improvement. Our experiments performed on three real world and one simulated data collections prove that the proposed quality model can effectively help improve the quality of the web-based imputation results by more than 15 percent, while our crowd cost saving strategy saves more than 75 percent crowd cost.

Abstract Online group buying is a burgeoning business model of Internet shopping, in which people with the same merchandise interests form a group and co-purchase goods with favorable prices. The buyer who launches the co-purchase is called the initiator, and other buyers are called the co-purchasers. Although recommending co-purchasers for a target buyer (co-purchase initiator) on the group buying is an interesting problem, existing studies have paid few attention to this topic. Different from the collaborator recommendation that only considers users with high similarity to the target user, co-purchaser recommendation takes both users with high and weak similarity into account, and the recommendation results can achieve high recall and diversity. However, the task turns out to be a challenging problem since it is hard to make a precise recommendation for buyers with weak similarity. To address the problem, we propose the following two methods. In the first one, we directly impose a penalty to the weak similar co-purchasers in the embedding space. To further improve the recommendation performance, in the second one, we smoothly increase the co-occurrence probability of the weak similar co-purchasers by truncated bias walk. Our experimental results on real datasets show that the proposed methods, particularly the latter, can effectively complete the co-purchaser recommendation and has high recommendation performance. In addition, considering that co-purchase may last longer, the total recommendation result can be generated in multiple stages and adjust the current recommendation list based on the feedback from the recommendation of previous stages. It is a trick for all co-purchaser recommendation methods to make the total result better.

In this paper, we study a new type of spatial crowdsourcing, namely competitive detour tasking, where workers can make detours from their original travel paths to perform multiple tasks, and each worker is allowed to compete for preferred tasks by strategically claiming his/her detour costs. The objective is to make suitable task assignment by maximizing the social welfare of crowdsourcing systems and protecting workers' private sensitive information. We first model the task assignment problem as a reverse auction process. We formalize the winning bid selection of reverse auction as an n-to-one weighted bipartite graph matching problem with multiple 0-1 knapsack constraints. Since this problem is NP-hard, we design an approximation algorithm to select winning bids and determine corresponding payments. Based on this, a Secure Reverse Auction (SRA) protocol is proposed for this novel spatial crowdsourcing. We analyze the approximation performance of the proposed protocol and prove that it has some desired properties, including truthfulness, individual rationality, computational efficiency, and security. To the best of our knowledge, this is the first theoretically provable secure auction protocol for spatial crowdsourcing systems. In addition, we also conduct extensive simulations on a real trace to verify the performance of the proposed protocol.

Next Point-of-Interest (POI) recommendation is of great value for both location-based service providers and users. However, the state-of-the-art Recurrent Neural Networks (RNNs) rarely consider the spatio-temporal intervals between neighbor check-ins, which are essential for modeling user check-in behaviors in next POI recommendation. To this end, in this paper, we propose a new Spatio-Temporal Gated Network (STGN) by enhancing long-short term memory network, where spatio-temporal gates are introduced to capture the spatio-temporal relationships between successive checkins. Specifically, two pairs of time gate and distance gate are designed to control the short-term interest and the longterm interest updates, respectively. Moreover, we introduce coupled input and forget gates to reduce the number of parameters and further improve efficiency. Finally, we evaluate the proposed model using four real-world datasets from various location-based social networks. The experimental results show that our model significantly outperforms the state-ofthe-art approaches for next POI recommendation.

In this demonstration, we present an end-to-end web-aided data imputation prototype system named WebPut. WebPut consults the Web for imputing the missing values in a local database when the traditional inferring-based imputation method has difficulties in getting the right answers. Specifically, WebPut investigates the interaction between the local inferring-based imputation methods and the web-based retrieving methods and shows that retrieving a small number of selected missing values can greatly improve the imputation recall of the inferring-based methods. Besides, WebPut also incorporates a crowd intervention component that can get advice from humans in case that the web-based imputation methods may have difficulties in making the right decisions. We demonstrate, step by step, how WebPut fills an incomplete table with each of its components.

As social animals, attending group activities is an indispensable part in people's daily social life, and it is an important task for recommender systems to suggest satisfying activities to a group of users. The major challenge in this task is how to aggregate personal preferences of group members to infer the decision of a group. Conventional group recommendation methods applied a predefined strategy for preference aggregation. However, these static strategies are too simple to model the real and complex process of group decision-making, especially for occasional groups which are formed ad-hoc. Moreover, group members should have non-uniform influences or weights in a group, and the weight of a user can be varied in different groups. Therefore, an ideal group recommender system should be able to accurately learn not only users' personal preferences but also the preference aggregation strategy from data. In this paper, we propose a novel group recommender system, namely SIGR (short for "Social Influence-based Group Recommender"), which takes an attention mechanism and a bipartite graph embedding model BGEM as building blocks. Specifically, we adopt an attention mechanism to learn each user's social influence and adapt their social influences to different groups and develop a novel deep social influence learning framework to exploit and integrate users' global and local social network structure information to further improve the estimation of users' social influences. BGEM is extended to model group-item interactions. In order to overcome the limitation and sparsity of the interaction data generated by occasional groups, we propose two model optimization approaches to seamlessly integrate the user-item interaction data. We create two large-scale benchmark datasets and conduct extensive experiments on them. The experimental results show the superiority of our proposed SIGR by comparing with state-of-the-art group recommender models.

An Attributed Dynamic Graph (ADG) contains multiple dynamic attributes associated with each edge. In ADG based applications, people usually can specify multiple constrains in the attributes to illustrate their requirements, such as the total cost, the total travel time and the stopover interval of a flight between two cities. This inspires a type of Multi-Constrained Temporal Path (MCTP) discovery in ADGs, which is a challenging NP-Complete problem. In order to deliver an efficient and effective temporal path discovery method to be used in real-time environment, we propose a Reinforcement Learning (RL) based, Monte Carlo Tree Search algorithm (RLMCTS). RL-MCTS uses a newly designed memory structure to address the challenges of Monte Carlo Tree Search (MCTS) in MCTP discovery. To the best of our knowledge, RL-MCTS is the first RL algorithm that supports path discovery in ADGs. The experimental results on ten real dynamic graphs demonstrate that our algorithm outperforms the state-of-the-art methods in terms of both efficiency and effectiveness.

Massive trajectory data stems from the prevalence of equipment supporting GPS and wireless communication technology. Based on these data, the computation of trajectory similarity has become a research hotspot in spatial database during recent years. The trajectory sampling problem caused by the different sampling strategies of the device has many negative effects on the similarity measurement. Although many recent studies have solved this problem by trajectory complements, these methods still have drawbacks because only spatial and temporal features are considered. Activity trajectory, with the development of LBSN (Location-based Social Network), endows traditional trajectory data with additional semantic information. In this paper, we fuse the spatio-temporal characteristics with extra activity information of the activity trajectory to solve the shortcomings in the process of trajectory complement. Specifically, we utilize vectors containing these three kinds of semantic information as the input of deep learning model for acquiring final trajectory representation, which is not only robust to low sampling, but also can capture implicit features in the trajectory. What's more, for the purpose of meeting the actual situation of network training, we propose a novel loss function based on the attention mechanism in natural language processing to distinguish these three kinds of information. Our framework, called At2vec, demonstrates better results than existing baselines when making extensive experiments on real trajectory databases.

The conventional methods for the next-item recommendation are generally based on RNN or one- dimensional attention with time encoding. They are either hard to preserve the long-term dependencies between different interactions, or hard to capture fine-grained user preferences. In this paper, we propose a Double Most Relevant Attention Network (DMRAN) that contains two layers, i.e., Item level Attention and Feature Level Self- attention, which are to pick out the most relevant items from the sequence of user’s historical behaviors, and extract the most relevant aspects of relevant items, respectively. Then, we can capture the fine-grained user preferences to better support the next-item recommendation. Extensive experiments on two real-world datasets illustrate that DMRAN can improve the efficiency and effectiveness of the recommendation compared with the state-of-the-art methods.

Question generation (QG) aims to create a fluent question from a passage and a target answer. State-of-the-art approaches are mainly based on encoder-decoder models to generate questions from the given passage and answer, which focus on using the information contained in a particular part of the passage for QG, but unaware of the clues hidden in other parts of the passage. Besides, the existing work on QG mainly focus on generating factoid questions, which are less suitable for generating non-factoid questions such as why-questions. In this paper, we propose to augment encoder-decoder framework with a pair-wise self-matching attention mechanism to dynamically collect inter-sentential evidence from the whole passage according to the current passage word and answer information. Besides, to let the model be more suitable for why-question generation, we also involve some causal features in the encoding process. Finally, to tackle the lack of why-question generation training data problem, we adopt a distant supervised method with an initial causal knowledge base to generate a large training data for why-question generation. Extensive experiments on several data sets show that our model significantly outperforms state-of-the-art question generation models not only on why-question generation tasks, but also on other types of question generation tasks.

Semantic drift is a common problem in iterative information extraction. Previous approaches for minimizing semantic drift may incur substantial loss in recall. We observe that most semantic drifts are introduced by a small number of questionable extractions in the earlier rounds of iterations. These extractions subsequently introduce a large number of questionable results, which lead to the semantic drift phenomenon. We call these questionable extractions Drifting Points (DPs). If erroneous extractions are the “symptoms” of semantic drift, then DPs are the “causes” of semantic drift. In this paper, we propose a method to minimize semantic drift by identifying the DPs and removing the effect introduced by the DPs. We use isA (concept-instance) extraction as an example to describe our approach in cleaning information extraction errors caused by semantic drift, but we perform experiments on different relation extraction processes on three large real data extraction collections. The experimental results show that our DP cleaning method enables us to clean around 90 percent incorrect instances or patterns with about 90 percent precision, which outperforms the previous approaches we compare with.

Graph Pattern Matching (GPM) has been used in lots of areas, like biology, medical science, and physics. With the advent of Online Social Networks (OSNs), recently, GPM has been playing a significant role in social network analysis, which has been widely used in, for example, finding experts, social community mining, and social position detection. Given a query which contains a pattern graph GQand a data graph GD, a GPM algorithm finds those subgraphs, GM, that match GQin GD. However, the existing GPM methods do not consider the multiple end-to-end constraints of the social contexts, like social relationships, social trust, and social positions on edges in GQ, which are commonly found in various applications, such as crowdsourcing travel, social network based ecommerce, and study group selection, etc. In this paper, we first conceptually extend Bounded Simulation to Multi-Constrained Simulation (MCS), and propose a novel NP-Complete Multi-Constrained Graph Pattern Matching (MC-GPM) problem. Then, to address the efficiency issue in large-scale MC-GPM, we propose a new concept called Strong Social Component (SSC), consisting of participants with strong social connections. We also propose an approach to identifying SSCs, and propose a novel index method and a graph compression method for SSC. Moreover, we devise a multithreading heuristic algorithm, called M-HAMC, to bidirectionally search the MC-GPM results in parallel without decompressing graphs. An extensive empirical study over five real-world large-scale social graphs has demonstrated the effectiveness and efficiency of our approach.

Information Extraction by Text Segmentation (IETS) aims at segmenting text inputs to extract implicit data values contained in them.The state-of-art IETS approaches mainly rely on machine learning techniques, either supervised or unsupervised.However, while the supervised approaches require a large labelled training data, the performance of the unsupervised ones could be unstable on different data sets.To overcome their weaknesses, this paper introduces CNN-IETS, a novel unsupervised probabilistic approach that takes the advantages of pre-existing data and a Convolution Neural Network (CNN)-based probabilistic classification model. While using the CNN model can ease the burden of selecting high-quality features in associating text segments with attributes of a given domain, the pre-existing data as a domain knowledge base can provide training data with a comprehensive list of features for building the CNN model.Given an input text, we do initial segmentation (according to the occurrences of these words in the knowledge base) to generate text segments for CNN classification with probabilities. Then, based on the probabilistic CNN classification results, we work on finding the most probable labelling way to the whole input text.As a complementary, a bidirectional sequencing model learned on-demand from test data is finally deployed to do further adjustment to some problematic labelled segments.Our experimental study conducted on several real data collections shows that CNN-IETS improves the extraction quality of state-of-art approaches by more than 10%.

Schema Matching (SM) and Record Matching (RM) are two necessary steps in integrating multiple relational tables of different schemas, where SM unifies the schemas and RM detects records referring to the same real-world entity. The two processes have been thoroughly studied separately, but few attention has been paid to the interaction of SM and RM. In this work we find that, even alternating them in a simple manner, SM and RM can benefit from each other to reach a better integration performance (i.e., in terms of precision and recall). Therefore, combining SM and RM is a promising solution for improving data integration.

Graph Pattern Matching (GPM) plays a significant role in many real applications, where given a graph pattern Q and a data graph G, computing the set M(Q, G) of matching subgraphs of Q in G. However, many applications like the experts recommendation in social networks, often need to find Top-K matches of a designated node v0, rather than the entire set M(Q, G). Moreover, the existing GPM method for matching the designated node v0does not consider the multiple constraints of the attributes associated with each vertex and each edge which commonly exist in real applications, like the constraints of social contexts for the experts recommendation in contextual social. In this paper, we first propose the Multi-Constrained Top-K Graph Pattern Matching problem (MC-Top-K-GPM), which involves the NP-Complete Multiple Constrained GPM problem. To address the efficiency and effectiveness issues of MC-Top-K-GPM in large-scale social graphs, we propose a novel index, called HB-Tree, which indexes the label and degree of nodes in G and can get candidates of v0efficiently. Furthermore, we develop a Multi-Constrained Top-K GPM method, called MTK, which can identify Top-K matches of v0effectively and efficiently. Using real-life data, we experimentally verify that MTK outperforms the existing GPM algorithms in both efficiency and effectiveness in solving the MC-TOP-K GPM problem.

Schema Matching (SM) and Record Matching (RM) are two necessary steps in integrating multiple relational tables of different schemas, where SM unifies the schemas and RM detects records referring to the same real-world entity. The two processes have been thoroughly studied separately, but few attention has been paid to the interaction of SM and RM. In this work, we find that, even alternating them in a simple manner, SM and RM can benefit from each other to reach a better integration performance (i.e., in terms of precision and recall). Therefore, combining SM and RM is a promising solution for improving data integration. To this end, we define novel matching rules for SM and RM, respectively, that is, every SM decision is made based on intermediate RM results, and vice versa, such that SM and RM can be performed alternately. The quality of integration is guaranteed by a Matching Likelihood Estimation model and the control of semantic drift, which prevent the effect of mismatch magnification. To reduce the computational cost, we design an index structure based on q-grams and a greedy search algorithm that can reduce around 90 percent overhead of the interaction. Extensive experiments on three data collections show that the combination and interaction between SM and RM significantly outperforms previous works that conduct SM and RM separately.

One of the fundamental questions is how we can enhance our relationships with the people in our social networks. The combination of frequent engagement, deep interaction, and time spent together are what constitute to build a stronger relationship. With this knowledge in mind, it is not surprising to see many scholars claim that social networking sites may help us to re-connect some lost relationships, but has limited ability to strengthen an existing relationship. Surprisingly, from a survey we conducted, we found that most people who use Facebook or scholat.com (a scholar-oriented social networking site) claim that their relationships with some of their friends are improved, even though they do not have any in-depth or personal conversation with them and they can understand their friends better even without any communication. It contradicts with what we have known. It seems that a relationship can be improved with Facebook or scholat.com alone. In this paper, we try to analyze this finding and explain it using communication model and visualized analysis. We conclude how communication model changes as time goes and how it re-defines the meaning of communication.

Data imputation aims at filling in missing attribute values in databases. Existing imputation approaches to nonquantitive string data can be roughly put into two categories: (1) inferring-based approaches [2], and (2) retrieving-based approaches [1]. Specifically, the inferring-based approaches find substitutes or estimations for the missing ones from the complete part of the data set. However, they typically fall short in filling in unique missing attribute values which do not exist in the complete part of the data set [1]. The retrieving-based approaches resort to external resources for help by formulating proper web search queries to retrieve web pages containing the missing values from the Web, and then extracting the missing values from the retrieved web pages [1]. This webbased retrieving approach reaches a high imputation precision and recall, but on the other hand, issues a large number of web search queries, which brings a large overhead [1].

In Online Social Networks (OSNs), the important participants, the trust relations between participants, and the interaction contexts between participants greatly impact a participant's decision-making in many applications, such as service provider selection and crowdsourcing service invocation. However, predicting the trust between two unknown participants based on the whole large-scale social network can lead to very high computation costs. Thus, prior to trust prediction, extracting a small-scale sub-network containing the important participants and the corresponding contextual information with a high density could make the trust prediction more efficient and effective. However, extracting such a sub-network has been proved to be an NP-Complete problem. To address this challenging problem, we propose a strong social component-aware trust sub-network extraction model, So-BiNet, to search for near-optimal solutions effectively and efficiently. Our method can extract a trust sub-network without any decompression, which can in turn greatly save the search time of trust sub-network extraction. The experiments, conducted on four social network datasets, demonstrate that our approach can efficiently extract sub-networks covering important participants and contextual information while keeping a high density. Our approach is superior to the state-of-the-art approaches in terms of the quality of the sub-networks extracted within the same execution time.

The problem we aim to address is the optimization of cost management for executing multiple continuous queries on data streams, where each query is defined by several filters, each of which monitors certain status of the data stream. Specially, the filter can be shared by different queries and expensive to evaluate. The conventional objective for such a problem is to minimize the overall execution cost to solve all queries, by planning the order of filter evaluation in shared strategy. However, in the streaming scenario, the characteristics of data items may change in process, which can bring some uncertainty to the outcome of individual filter evaluation, and affect the plan of query execution as well as the overall execution cost. In our work, considering the influence of the uncertain variation of data characteristics, we propose a framework to deal with the dynamic adjustment of filter ordering for query execution on data stream, and focus on the issues of cost management. By incrementally monitoring and analyzing the results of filter evaluation, our proposed approach can be effectively adaptive to the varied stream behavior and adjust the optimal ordering of filter evaluation, so as to optimize the execution cost. In order to achieve satisfactory performance and efficiency, we also discuss the trade-off between the adaptivity of our framework and the overhead incurred by filter adaption. The experimental results on synthetic and two real data sets (traffic and multimedia) show that our framework can effectively reduce and balance the overall query execution cost and keep high adaptivity in streaming scenario.

Online Social Networks (OSNs) have been used as the means for a variety of applications, like employment system, e-Commerce and CRM system. In these applications, social influence acts as a significant role, affecting people's decision-making. However, the existing social influence evaluation methods do not fully consider the social contexts, like the social relationships and the social trust between participants, and the preferences of participants, which have significant impact on social influence evaluation in OSNs. Thus, these existing methods cannot deliver accurate social influence evaluation results. In our paper, we propose a Context-Aware Trust-Oriented Influencers Finding method, called CT-Influence, with social contexts taken into account. We conduct experiments onto two real social network datasets, i.e., Epinions and DBLP. The experimental results illustrate that our CT-Influence method greatly outperforms the state-of-the-art method So Cap in terms of effectiveness and efficiency.

Web Harvesting enables the enrichment of incomplete data sets by retrieving required information from the Web. However, the ambiguity of instances may greatly decrease the quality of the harvested data, given that any instance in the local data set may become ambiguous when attempting to identify it on the Web. Although plenty of disambiguation methods have been proposed to deal with the ambiguity problems in various settings, none of them are able to handle the instance ambiguity problem in Web Harvesting. In this paper, we propose to do instance disambiguation in Web Harvesting with a novel disambiguation method inspired by the idea of collaborative identity recognition. In particular, we expect to find some common properties in forms of latent shared attribute values among instances in the list, such that these shared attribute values can differentiate instances within the list against those ambiguous ones on the Web. Our extensive experimental evaluation illustrates the utility of collaborative disambiguation for a popular Web Harvesting application, and shows that it substantially improves the accuracy of the harvested data.

Data imputation aims at filling in missing attribute values in databases. Most existing imputation methods to string attribute values are inferring-based approaches, which usually fail to reach a high imputation recall by just inferring missing values from the complete part of the data set. Recently, some retrieving-based methods are proposed to retrieve missing values from external resources such as the World Wide Web, which tend to reach a much higher imputation recall, but inevitably bring a large overhead by issuing a large number of search queries. In this paper, we investigate the interaction between the inferring-based methods and the retrieving-based methods. We show that retrieving a small number of selected missing values can greatly improve the imputation recall of the inferring-based methods. With this intuition, we propose an inTeractive Retrieving-Inferring data imPutation approach (TRIP), which performs retrieving and inferring alternately in filling in missing attribute values in a data set. To ensure the high recall at the minimum cost, TRIP faces a challenge of selecting the least number of missing values for retrieving to maximize the number of inferable values. Our proposed solution is able to identify an optimal retrieving-inferring scheduling scheme in deterministic data imputation, and the optimality of the generated scheme is theoretically analyzed with proofs. We also analyze with an example that the optimal scheme is not feasible to be achieved in τ-constrained stochastic data imputation (τ-SDI), but still, our proposed solution identifies an expected-optimal scheme in τ-SDI. Extensive experiments on four data collections show that TRIP retrieves on average 20 percent missing values and achieves the same high recall that was reached by the retrieving-based approach.

Semantic drift is a common problem in iterative information extraction. Previous approaches for minimizing semantic drift may incur substantial loss in recall. We observe that most semantic drifts are introduced by a small number of questionable extractions in the earlier rounds of iterations. These extractions subsequently introduce a large number of questionable results, which lead to the semantic drift phenomenon. We call these questionable extractions Drifting Points (DPs). If erroneous extractions are the "symptoms" of semantic drift, then DPs are the "causes" of semantic drift. In this paper, we propose a method to minimize semantic drift by identifying the DPs and removing the effect introduced by the DPs. We use isA (concept-instance) extraction as an example to demonstrate the effectiveness of our approach in cleaning information extraction errors caused by semantic drift. We perform experiments on a isA relation iterative extraction, where 90.5 million of isA pairs are automatically extracted from 1.6 billion web documents with a low precision. The experimental results show our DP cleaning method enables us to clean more than 90% incorrect instances with 95% precision, which outperforms the previous approaches we compare with. As a result, our method greatly improves the prevision of this large isA data set from less than 50% to over 90%.

We identify Relation Completion (RC) as one recurring problem that is central to the success of novel big data applications such as Entity Reconstruction and Data Enrichment.Given a semantic relation R, RC attempts at linking entity pairs between two entity lists under the relation R. To accomplish the RC goals, we propose to formulate search queries for each query entity α based on some auxiliary information, so that to detect its target entity β from the set of retrieved documents.For instance, a Pattern-based method (PaRE) uses extracted patterns as the auxiliary information in formulating search queries.However, high-quality patterns may decrease the probability of finding suitable target entities.As an alternative, we propose CoRE method that uses context terms learned surrounding the expression of a relation as the auxiliary information in formulating queries.The experimental results based on several real-world web data collections demonstrate that CoRE reaches a much higher accuracy than PaRE for the purpose of RC.

In this paper, we propose a new type of Dictionary-based Entity Recognition Problem, named Approximate Membership Localization (AML). The popular Approximate Membership Extraction (AME) provides a full coverage to the true matched substrings from a given document, but many redundancies cause a low efficiency of the AME process and deteriorate the performance of real-world applications using the extracted substrings. The AML problem targets at locating nonoverlapped substrings which is a better approximation to the true matched substrings without generating overlapped redundancies. In order to perform AML efficiently, we propose the optimized algorithm P-Prune that prunes a large part of overlapped redundant matched substrings before generating them. Our study using several real-word data sets demonstrates the efficiency of P-Prune over a baseline method. We also study the AML in application to a proposed web-based join framework scenario which is a search-based approach joining two tables using dictionary-based entity recognition from web documents. The results not only prove the advantage of AML over AME, but also demonstrate the effectiveness of our search-based approach.

In a pilot application based on web search engine called Web-based Relation Completion (WebRC), we propose to join two columns of entities linked by a predefined relation by mining knowledge from the web through a web search engine. To achieve this, a novel retrieval task Relation Query Expansion (RelQE) is modelled: given an entity (query), the task is to retrieve documents containing entities in predefined relation to the given one. Solving this problem entails expanding the query before submitting it to a web search engine to ensure that mostly documents containing the linked entity are returned in the top K search results. In this paper, we propose a novel Learning-based Relevance Feedback (LRF) approach to solve this retrieval task. Expansion terms are learned from training pairs of entities linked by the predefined relation and applied to new entity-queries to find entities linked by the same relation. After describing the approach, we present experimental results on real-world web data collections, which show that the LRF approach always improves the precision of top-ranked search results to up to 8.6 times the baseline. Using LRF, WebRC also shows performances way above the baseline.

In this paper, we propose a search-based approach to join two tables in the absence of clean join attributes. Non-structured documents from the web are used to express the correlations between a given query and a reference list. To implement this approach, a major challenge we meet is how to efficiently determine the number of times and the locations of each clean reference from the reference list that is approximately mentioned in the retrieved documents. We formalize the Approximate Membership Localization (AML) problem and propose an efficient partial pruning algorithm to solve it. A study using real-word data sets demonstrates the effectiveness of our search-based approach, and the efficiency of our AML algorithm.

Influence between objects needs to be assessed in many applications. Lots of measures have been proposed, but a domain-independent method is still expected. In this paper, we give a probabilistic definition of influence based on the random walker model on graphs. Two approaches, linear systems method and Basic InfRank algorithm, are shown and return equal results, but Basic InfRank is more efficient by iterative computation. Two variants on bipartite graphs and star graphs are discussed. Experiments show InfRank algorithms have good accuracy, fast convergent rate and high performance.