论文检索

Zhengzhong Zhu, Pei Zhou, Zeting Li, Kejiang Chen, Jiangping Zhu

Multi-label text classification (MLTC) is a challenging task where each document can be associated with multiple interdependent labels. This task is complicated by two key issues: the intricate correlations among labels and the partial overlap between labels and text relevance. Existing methods often fail to capture the semantic dependencies between labels or struggle to handle the ambiguities caused by partial overlaps, resulting in suboptimal representation learning. To address these challenges, we propose the Unified Contextual and Label-Aware Framework (UCLAF), which integrates a Label Attention Aware Network(LAN) and Correlation Aware Contrastive Learning (CACL) in a synergistic design. The Label Attention Aware Network explicitly models label dependencies by embedding labels and texts into a shared semantic space, aligning text representations with label semantics. Meanwhile, Correlation Aware Contrastive Learning refines these representations by dynamically modeling sample-level relationships, leveraging a contrastive loss function that accounts for the proportional overlap of labels between samples. This complementary approach enables UCLAF to jointly address complex label correlations and partial label overlaps. Extensive experiments on benchmark datasets demonstrate that UCLAF significantly outperforms state-of-the-art methods, showcasing its effectiveness in improving both representation learning and classification performance in MLTC tasks. We will release our code after the paper is accepted.

Linlin Zhu, Heli Sun, Xiaoyong Huang, Qi Zhang, Ruichen Cao, Liang He

Multimodal aspect-based sentiment analysis aims to extract aspects from different data sources and recognize the corresponding sentiments. While current research has broadly focused on syntax relation-driven semantic comprehension, the impact of the importance of different syntactic relations on semantic understanding has not been adequately investigated. To address this issue, we propose a Sentiment-enhanced Multi-hop Connected Graph Attention Network (MCG), aiming to enhance the discriminative capability of model for sentiments and to delve into the syntactic relationships within the text. Firstly, we design a contrastive sentiment-enhanced pre-training task that expands the diversity and complexity of training samples to improve the recognition of multiple sentiments. Secondly, we construct a multi-hop connected syntactic dependency graph to deeply explore the rich syntactic dependencies in the text and to reveal the differences among syntactic relations. Moreover, we develop a multi-hop connected graph attention mechanism that enables the model to focus on the key syntactic relations within the syntactic structure, thereby enhancing the comprehension and predictive capabilities of model in multimodal sentiment analysis. Experimental results on two benchmark datasets demonstrate that our method outperforms state-of-the-art methods. The source code is provided in the supplementary materials.

Miao Zhang, Jiawei Wang, Jinying Han, Kui Xiao, Zhifei Li, Yan Zhang, Hao Chen, Shihui Wang

Concept prerequisite relations determine the learning order of knowledge concepts in one domain, which has an important impact on teachers' course design and students' personalized learning. Current research usually predicts concept prerequisite relations from the perspective of knowledge, and rarely pays attention to the role of learners' learning behavior. We propose a Dual Graph Distillation Method for Concept Prerequisite Relation Learning (DGCPL). Specifically, DGCPL constructs a dual graph structure from both the knowledge and learning behavior perspectives, and captures the high-order knowledge features and learning behavior features through the concept-resource hypergraph and the learning behavior graph respectively. In addition, we introduce a gated knowledge distillation to fuse the structural information of concept nodes in the two graphs, so as to obtain a more comprehensive concept embedding representation and achieve accurate prediction of prerequisite relations. On three public benchmark datasets, we compare DGCPL with eight graph-based baseline methods and five traditional classification baseline methods. The experimental results show that DGCPL achieves state-of-the-art performance in learning concept prerequisite relations. Our code is available at https://github.com/wisejw/DGCPL.

Xinyu Yao, Mengdi Wang, Bo Chen, Xiaobing Zhao

Classical Chinese, as the core carrier of Chinese culture, plays a crucial role in the inheritance and study of ancient literature. However, existing natural language processing models primarily optimize for Modern Chinese, resulting in inadequate performance on Classical Chinese. This paper presents a comprehensive solution for Classical Chinese language processing. By continuing pre-training and instruction fine-tuning on the LLaMA3-8B-Chinese model, we construct a large language model, WenyanGPT, which is specifically designed for Classical Chinese tasks. Additionally, we develop an evaluation benchmark dataset, WenyanBENCH. Experimental results on WenyanBENCH demonstrate that WenyanGPT significantly outperforms current advanced LLMs in various Classical Chinese tasks. We make the model's training data, instruction fine-tuning data, and evaluation benchmark dataset publicly available to promote further research and development in the field of Classical Chinese processing.

Bo Xu, Liangzhi Li, Junlong Wang, Xuening Qiao, Erchen Yu, Yiming Qian, Linlin Zong, Hongfei Lin

The rapid development of large language models has greatly advanced human-computer dialogue research. However, applying these models to specialized fields like maternity and infant care often leads to subpar performance due to a lack of domain-specific datasets. To address this problem, we have created MicDialogue, a Chinese dialogue dataset for maternity and infant care. MicDialogue involves a wide range of specialized topics, including gynecological health, pediatric care, pregnancy preparation, emotional counseling and other related topics. This dataset is curated from two types of Chinese social media: short videos and blog posts. Short videos capture real-time interactions and pragmatic dialogue patterns, while blog posts offer comprehensive coverage of various topics within the domain. We have also included detailed annotations for topics, diseases, symptoms, and causes, enabling in-depth research. Additionally, we developed a knowledge-driven benchmark model using LLM-based prompt learning and multiple knowledge graphs to address diverse dialogue topics. Experiments validate MicDialogue's usability, providing benchmarks for future research and essential data for fine-tuning language models in maternity and infant care.

Yifan Wei, Xiaoyan Yu, Ran Song, Hao Peng, Angsheng Li

Large Language Models (LLMs) excel in tasks such as retrieval and question answering but require updates to incorporate new knowledge and reduce inaccuracies and hallucinations. Traditional updating methods, like fine-tuning and incremental learning, face challenges such as overfitting and high computational costs. Knowledge Editing (KE) provides a promising alternative but often overlooks the Knowledge Element Overlap (KEO) phenomenon, where multiple triplets share common elements, leading to editing conflicts. We identify the prevalence of KEO in existing KE datasets and show its significant impact on current KE methods, causing performance degradation in handling such triplets. To address this, we propose a new formulation, Knowledge Set Editing (KSE), and introduce SetKE, a method that edits sets of triplets simultaneously. Experimental results demonstrate that SetKE outperforms existing methods in KEO scenarios on mainstream LLMs. Additionally, we introduce EditSet, a dataset containing KEO triplets, providing a comprehensive benchmark.

Peng Wang, Shuo Li, Ruoxi Zhou, Qiguang Chen, Xiao Xu, Hao Fei, Dagang Li, Wanxiang Che, Libo Qin

Consistency identification in task-oriented dialog (CI-ToD) typically consists of three sub-tasks: User Query Inconsistency (QI) identification, Dialogue History Inconsistency (HI) identification, and Knowledge Base Inconsistency (KBI) identification, which aim to determine inconsistent relationships between system response and user query, dialogue history, and knowledge base. Previous approaches focus on the exploration of deep learning models for CI-ToD. While these models achieve remarkable progress, they still rely on large amounts of labeled data, which is hard to achieve in real-world scenarios. Motivated by this, in the paper, we aim to explore large language models for CI-ToD, which do not require any training data. In addition, we further introduce a multi-agent collaboration framework (MAC-CIToD) to model the interaction across three sub-tasks in CI-ToD, including (1) Full Connection paradigm, (2) Cycle Connection paradigm, and (3) Central Connection paradigm, which effectively builds interaction across QI, HI, and KBI. Experiments on the standard benchmark reveal that our framework achieves superior performance. Additionally, we compare MAC-CIToD with the most advanced trained approaches and find that its zero-shot performance on most metrics even surpasses that of models after training on the CI-ToD dataset.

Behzad Mehrbakhsh, Fernando Martínez-Plumed, José Hernández-Orallo

Contamination in large language models (LLMs), and machine learning more broadly, refers to the inclusion of equal --or very similar-- examples in both training and test sets. This phenomenon usually translates into better test performance. Here we explore when this contamination is performed intentionally, for purposes that can be malicious (e.g., get better scores in evaluations) or benevolent (e.g., fix some mistakes). These interventions, usually in the form of fine-tuning memorisations, come with a budget in the size of the fine-tuning dataset. Several trade-offs appear between the breadth of the intervention (how many examples to be memorised), its depth (how many repetitions of each example) and the difficulty of the examples. By studying several LLMs and datasets, we observe some monotonic behaviour (more difficult items require more depth to be `fixed') but also some non-monotonic phenomena (very high depth levels have negative effects on non-contaminated examples). This suggests that trade-offs should be found not only in terms of the budget but also according to model specifics, the task and the item difficulty at hand.

Qian Li, Cheng Ji, Shu Guo, Kun Peng, Qianren Mao, Shangguang Wang

Multi-modal relation extraction (MMRE) is a challenging task that seeks to identify relationships between entities with textual and visual attributes. However, existing methods struggle to handle the complexities posed by multiple entity pairs within a single sentence that share similar contextual information (e.g., identical text and image content). These scenarios amplify the difficulty of distinguishing relationships and hinder accurate extraction. To address these limitations, we propose the variational multi-modal hypergraph attention network (VM-HAN), a novel and robust framework for MMRE. Unlike previous approaches, VM-HAN constructs a multi-modal hypergraph for each sentence-image pair, explicitly modeling high-order intra-/inter-modal correlations among different entity pairs in the same context. This design enables a more detailed and nuanced understanding of entity relationships by capturing intricate cross-modal interactions that are often overlooked. Additionally, we introduce the variational hypergraph attention network (V-HAN). This variational attention mechanism dynamically refines the hypergraph structure, enabling the model to effectively handle the inherent ambiguity and complexity of multi-modal data. Comprehensive experiments on benchmark MMRE datasets demonstrate that VM-HAN achieves state-of-the-art performance, significantly surpassing existing methods in both accuracy and efficiency.

Manzong Huang, Chenyang Bu, Yi He, Xindong Wu

Knowledge Graph (KG)-augmented Large Language Models (LLMs) have recently propelled significant advances in complex reasoning tasks, thanks to their broad domain knowledge and contextual awareness. Unfortunately, current methods often assume KGs to be complete, which is impractical given the inherent limitations of KG construction and the potential loss of contextual cues when converting unstructured text into entity-relation triples. In response, this paper proposes the Triple Context Restoration and Query-driven Feedback (TCR-QF) framework, which reconstructs the textual context underlying each triple to mitigate information loss, while dynamically refining the KG structure by iteratively incorporating query-relevant missing knowledge. Experiments on five benchmark question-answering datasets substantiate the effectiveness of TCR-QF in KG and LLM integration, where itachieves a 29.1% improvement in Exact Match and a 15.5% improvement in F1 over its state-of-the-art GraphRAG competitors. The code is publicly available at https://github.com/HFUT-DMiC-Lab/TCR-QF.git.

Quanjiang Guo, Jinchuan Zhang, Sijie Wang, Ling Tian, Zhao Kang, Bin Yan, Weidong Xiao

Few-Shot Relation Extraction (FSRE) remains a challenging task due to the scarcity of annotated data and the limited generalization capabilities of existing models. Although large language models (LLMs) have shown potential in FSRE through in-context learning, their general-purpose training objectives often result in suboptimal performance for task-specific relation extraction. To overcome these challenges, we propose TKRE (Two-Stage Knowledge-Guided Pre-training for Relation Extraction), a novel framework that synergistically integrates LLMs with traditional relation extraction models, bridging generative and discriminative learning paradigms. TKRE introduces two key innovations: (1) leveraging LLMs to generate explanation-driven knowledge and schema-constrained synthetic data, addressing the issue of data scarcity; and (2) a two-stage pre-training strategy combining Masked Span Language Modeling (MSLM) and Span-Level Contrastive Learning (SCL) to enhance relational reasoning and generalization. Together, these components enable TKRE to effectively handle FSRE tasks. Comprehensive experiments on benchmark datasets demonstrate the efficacy of TKRE, achieving new state-of-the-art performance in FSRE and underscoring its potential for broader application in low-resource scenarios. The code and data are released on https://github.com/UESTC-GQJ/TKRE.

Bingzhe Zhou, Hannan Wang, Yuan Yao, Taolue Chen, Feng Xu, Xiaoxing Ma

Satisfiability Modulo Theories (SMT) solvers are crucial in many applications, yet their performance is often a bottleneck. This paper introduces SIRISMT, a novel framework that employs machine learning techniques for the automatic synthesis of efficient SMT-solving strategies. Specifically, SIRISMT targets at Z3 and consists of three key stages. First, given a set of training SMT formulas, SIRISMT simulates the solving process by leveraging reinforcement learning to guide its exploration within the strategy space. Next, SIRISMT refines the collected strategies by pruning redundant tactics and generating augmented strategies based on the subsequence structure of the learned strategies. These refined strategies are then fed back into the reinforcement learning model. Finally, the refined and optimized strategies are integrated into one strategy, which can be directly plugged into modern SMT solvers. Extensive evaluations show the superior performance of SIRISMT over the baseline methods. For example, compared to the default Z3, it solves 26.8% more formulas and achieves up to an 86.3% improvement in the Par-2 score on benchmark datasets. Additionally, we show that the synthesized strategy can improve the code coverage by up to 11.8% in a downstream symbolic execution benchmark.

Tianlin Zhang, En Yu, Yi Shao, Jiande Sun

Multimodal fake news detection has garnered significant attention due to its profound implications for social security. While existing approaches have contributed to understanding cross-modal consistency, they often fail to leverage modal-specific representations and explicit discrepant features. To address these limitations, we propose a Multimodal Inverse Attention Network (MIAN), a novel framework that explores intrinsic discriminative features based on news content to advance fake news detection. Specifically, MIAN introduces a hierarchical learning module that captures diverse intra-modal relationships through local-to-global and local-to-local interactions, thereby generating enhanced unimodal representations to improve the identification of fake news at the intra-modal level. Additionally, a cross-modal interaction module employs a co-attention mechanism to establish and model dependencies between the refined unimodal representations, facilitating seamless semantic integration across modalities. To explicitly extract inconsistency features, we propose an inverse attention mechanism that effectively highlights the conflicting patterns and semantic deviations introduced by fake news in both intra- and inter-modality. Extensive experiments on benchmark datasets demonstrate that MIAN significantly outperforms state-of-the-art methods, underscoring its pivotal contribution to advancing social security through enhanced multimodal fake news detection.

Ning Wang, Zihan Yan, Weiyang Li, Chuan Ma, He Chen, Tao Xiang

Embodied agents exhibit immense potential across a multitude of domains, making the assurance of their behavioral safety a fundamental prerequisite for their widespread deployment. However, existing research predominantly concentrates on the security of general large language models, lacking specialized methodologies for establishing safety benchmarks and input moderation tailored to embodied agents. To bridge this gap, this paper introduces a novel input moderation framework, meticulously designed to safeguard embodied agents. This framework encompasses the entire pipeline, including taxonomy definition, dataset curation, moderator architecture, model training, and rigorous evaluation. Notably, we introduce EAsafetyBench, a meticulously crafted safety benchmark engineered to facilitate both the training and stringent assessment of moderators specifically designed for embodied agents. Furthermore, we propose Pinpoint, an innovative prompt-decoupled input moderation scheme that harnesses a masked attention mechanism to effectively isolate and mitigate the influence of functional prompts on moderation tasks. Extensive experiments conducted on diverse benchmark datasets and models validate the feasibility and efficacy of the proposed approach. The results demonstrate that our methodologies achieve an impressive average detection accuracy of 94.58%, surpassing the performance of existing state-of-the-art techniques, alongside an exceptional moderation processing time of merely 0.002 seconds per instance. The source code and datasets can be found at https://github.com/ZihanYan-CQU/EAsafetyBench.

Bing Wang, Ximing Li, Changchun Li, Bingrui Zhao, Bo Fu, Renchu Guan, Shengsheng Wang

The development of Internet technology has led to an increased prevalence of misinformation, causing severe negative effects across diverse domains. To mitigate this challenge, Misinformation Detection (MD), aiming to detect online misinformation automatically, emerges as a rapidly growing research topic in the community. In this paper, we propose a novel plug-and-play augmentation method for the MD task, namely Misinformation Detection with Potential Commonsense Conflict (MD-PCC). We take inspiration from the prior studies indicating that fake articles are more likely to involve commonsense conflict. Accordingly, we construct commonsense expressions for articles, serving to express potential commonsense conflicts inferred by the difference between extracted commonsense triplet and golden ones inferred by the well-established commonsense reasoning tool COMET. These expressions are then specified for each article as augmentation. Any specific MD methods can be then trained on those commonsense-augmented articles. Besides, we also collect a novel commonsense-oriented dataset named CoMis, whose all fake articles are caused by commonsense conflict. We integrate MD-PCC with various existing MD backbones and compare them across both 4 public benchmark datasets and CoMis. Empirical results demonstrate that MD-PCC can consistently outperform the existing MD baselines.

Xiangxiang Shen, Zheng Wan, Lingfeng Wen, Licheng Sun, Jian Yang, Xuan Tang, Shing-Ho J. Lin, Xiao He, Mingsong Chen, Xian Wei

Crystal structures can be simplified as a periodic point set that repeats across three-dimensional space along an underlying lattice. Traditionally, crystal representation methods rely on descriptors such as lattice parameters, symmetry, and space groups to characterize the structure. However, in reality, atoms in materials always vibrate above absolute zero, causing their positions to fluctuate continuously. This dynamic behavior disrupts the fundamental periodicity of the lattice, making crystal graphs based on static lattice parameters and conventional descriptors discontinuous under slight perturbations. Chemists proposed the pairwise distance distribution (PDD) method to address this. However, the completeness of PDD requires defining a large number of neighboring atoms, leading to high computational costs. Additionally, PDD does not account for atomic information, making it challenging to apply it directly to crystal material property prediction tasks. To tackle these challenges, we introduce the atom-weighted Pairwise Distance Distribution (WPDD) and Unit cell Pairwise Distance Distribution (UPDD) for the first time, applying them to the construction of multi-edge crystal graphs. We demonstrate the continuity and general completeness of crystal graphs under slight atomic position perturbations. Moreover, by modeling PDD as global information and integrating it into matrix-based message passing, we significantly reduce computational costs. Comprehensive evaluation results show that WPDDFormer achieves state-of-the-art predictive accuracy across tasks on benchmark datasets such as the Materials Project and JARVIS-DFT.

Minwoo Oh, Minsu Park, Eunil Park

Short video platforms like YouTube Shorts and TikTok face significant copyright compliance challenges, as infringers frequently embed arbitrary background music (BGM) to obscure original soundtracks (OST) and evade content originality detection. To tackle this issue, we propose a novel pipeline that integrates Music Source Separation (MSS) and cross-modal video-music retrieval (CMVMR). Our approach effectively separates arbitrary BGM from the original OST, enabling the restoration of authentic video audio tracks. To support this work, we introduce two domain-specific datasets: OASD-20K for audio separation and OSVAR-160 for pipeline evaluation. OASD-20K contains 20,000 audio clips featuring mixed BGM and OST pairs, while OSVAR160 is a unique benchmark dataset comprising 1,121 video and mixed-audio pairs, specifically designed for short video restoration tasks. Experimental results demonstrate that our pipeline not only removes arbitrary BGM with high accuracy but also restores OSTs, ensuring content integrity. This approach provides an ethical and scalable solution to copyright challenges in user-generated content on short video platforms.

Dung Nguyen, Minh Khoi Ho, Huy Ta, Thanh Tam Nguyen, Qi Chen, Kumar Rav, Quy Duong Dang, Satwik Ramchandre, Son Lam Phung, Zhibin Liao 等

Medical Large Multi-modal Models (LMMs) have demonstrated remarkable capabilities in medical data interpretation. However, these models frequently generate hallucinations contradicting source evidence, particularly due to inadequate localization reasoning. This work reveals a critical limitation in current medical LMMs: instead of analyzing relevant pathological regions, they often rely on linguistic patterns or attend to irrelevant image areas when responding to disease-related queries. To address this, we introduce HEAL-MedVQA (Hallucination Evaluation via Localization MedVQA), a comprehensive benchmark designed to evaluate LMMs' localization abilities and hallucination robustness. HEAL-MedVQA features (i) two innovative evaluation protocols to assess visual and textual shortcut learning, and (ii) a dataset of 67K VQA pairs, with doctor-annotated anatomical segmentation masks for pathological regions. To improve visual reasoning, we propose the Localize-before-Answer (LobA) framework, which trains LMMs to localize target regions of interest and self-prompt to emphasize segmented pathological areas, generating grounded and reliable answers. Experimental results demonstrate that our approach significantly outperforms state-of-the-art biomedical LMMs on the challenging HEAL-MedVQA benchmark, advancing robustness in medical VQA.

Hongyi Lyu, Xuyun Zhang, Hongsheng Hu, Shuo Wang, Chaoxiang He, Lianyong Qi

As machine learning models become widely deployed in data-driven applications, ensuring compliance with the 'right to be forgotten' as required by many privacy regulations is vital for safeguarding user privacy. To forget the given data, existing re-labeling based unlearning methods employ a single-step adjustment scheme that revises the decision boundaries in one re-labeling phase. However, such single-step approaches lead to coarse-grained changes in decision boundaries among the remaining classes and impose adverse effects on the model utility. To address these limitations, we propose 'Self-Unlearning with Layered Iteration (SULI),' a novel unlearning approach that introduces a layered iteration strategy to re-label the forgetting data iteratively and refine the decision boundaries progressively. We further develop a 'Selective Probability Adjustment (SPA)' technique, which uses a soft-label mechanism to promote smoother decision-boundary transitions. Comprehensive experiments on three benchmark datasets demonstrate that SULI achieves superior performance in effectiveness, efficiency, and privacy compared to the state-of-the-art baselines in both class-wise and instance-wise unlearning scenarios. The source code is released at https://github.com/Hongyi-Lyu-MQ/SULI.

Xiaoling Luo, Peng Chen, Chengliang Liu, Xiaopeng Jin, Jie Wen, Yumeng Liu, Junsong Wang

Multimodal protein features play a crucial role in protein function prediction. However, these features encompass a wide range of information, ranging from structural data and sequence features to protein attributes and interaction networks, making it challenging to decipher their complex interconnections. In this work, we propose a multimodal protein function prediction method (DSRPGO) by utilizing dynamic selection and reconstructive pre-training mechanisms. To acquire complex protein information, we introduce reconstructive pre-training to mine more fine-grained information with low semantic levels. Moreover, we put forward the Bidirectional Interaction Module (BInM) to facilitate interactive learning among multimodal features. Additionally, to address the difficulty of hierarchical multi-label classification in this task, a Dynamic Selection Module (DSM) is designed to select the feature representation that is most conducive to current protein function prediction. Our proposed DSRPGO model improves significantly in BPO, MFO, and CCO on human datasets, thereby outperforming other benchmark models.