论文检索

Fanxiao Li, Jiaying Wu, Tingchao Fu, Dayang Li, Herun Wan, Wei Zhou, Min-Yen Kan

Even when factually correct, social-media news previews (image-headline pairs) can induce interpretation drift: by selectively omitting crucial context, they lead readers to form judgments that diverge from what the full article supports. This covert harm is subtler than explicit misinformation, yet remains underexplored. To address this gap, we develop a multi-stage pipeline that simulates preview-based and context-based understanding, enabling construction of the MM-Misleading benchmark. Using MM-Misleading, we systematically evaluate open-source LVLMs and uncover pronounced blind spots in omission-based misleadingness detection. We further propose OMGuard, which combines (1) Interpretation-Aware Fine-Tuning for misleadingness detection and (2) Rationale-Guided Misleading Content Correction, where explicit rationales guide headline rewriting to reduce misleading impressions. Experiments show that OMGuard lifts an 8B model’s detection accuracy to the level of a 235B LVLM while delivering markedly stronger end-to-end correction. Further analysis shows that misleadingness usually arises from local narrative shifts, such as missing background, instead of global frame changes, and identifies image-driven cases where text-only correction fails, underscoring the need for visual interventions.

Ziyi Wang, Chen Zhang, Wenjun Peng, Qi Wu, Xinyu Wang

Explainability for Large Language Model (LLM) agents is especially challenging in interactive, partially observable settings, where decisions depend on evolving beliefs and other agents. We present TriEx, a tri-view explainability framework that instruments sequential decision making with aligned artifacts: (i) structured first-person self-reasoning bound to an action, (ii) explicit second-person belief states about opponents updated over time, and (iii) third-person oracle audits grounded in environment-derived reference signals. This design turns explanations from free-form narratives into evidence-anchored objects that can be compared and checked across time and perspectives. Using imperfect-information strategic games as a controlled testbed, we show that TriEx enables scalable analysis of explanation faithfulness, belief dynamics, and evaluator reliability, revealing systematic mismatches between what agents say, what they believe, and what they do. Our results highlight explainability as an interaction-dependent property and motivate multi-view, evidence-grounded evaluation for LLM agents.

I-Fan Lin, Faegheh Hasibi, Suzan Verberne

In this paper, we propose a training-free method for unsupervised short text clustering that relies less on careful selection of embedders than other methods. In customer-facing chatbots, companies are dealing with large amounts of user utterances that need to be clustered according to their intent. In these settings, no labeled data is typically available, and the number of clusters is not known. Recent approaches to short-text clustering in label-free settings incorporate LLM output to refine existing embeddings. While LLMs can identify similar texts effectively, the resulting similarities may not be directly represented by distances in the dense vector space, as they depend on the original embedding. We therefore propose a method for transforming LLM judgments directly into a bag-of-texts representation in which texts are initialized to be equidistant, without assuming any prior distance relationships. Our method achieves comparable or superior results to state-of-the-art methods, but without embeddings optimization or assuming prior knowledge of clusters or labels. Experiments on diverse datasets and smaller LLMs show that our method is model agnostic and can be applied to any embedder, with relatively small LLMs, and different clustering methods. We also show how our method scales to large datasets, reducing the computational cost of the LLM use. The flexibility and scalability of our method make it more aligned with real-world training-free scenarios than existing clustering methods.

Junhao Jia, Hanwen Zheng, Yueyi Wu, Huangwei Chen, Haishuai Wang, Jiajun Bu, Lei Wu

Natural language data is inherently noisy, yet standard interpretable models often rely on scalar similarities that obscure the true evidentiary basis of a prediction. This limitation is particularly detrimental to prototype-based classification, where traditional full-alignment mechanisms force non-informative background segments to match informative prototypes, yielding unstable or misleading explanations. To mitigate this, we present SCOUT, a novel paradigm that grounds prototype reasoning in the selective correspondence of discriminative fragments. Concretely, we represent each document as a discrete distribution over span embeddings and employ differentiable Unbalanced Optimal Transport (UOT) to align them with class-specific prototypes. Unlike standard methods, this mechanism enables the model to focus strictly on decisive evidence while leaving irrelevant noise unmatched via geometric mass suppression. To ensure verifiability, we anchor prototype supports to readable training spans, establishing a transparent bridge between input segments and stored knowledge. Comprehensive experiments on seven benchmarks demonstrate that SCOUT yields prototypes focused on semantically significant spans, significantly outperforming traditional rationale extraction and post-hoc attribution methods in terms of faithfulness and stability.

Mutaz Ayesh, Saif M. Mohammad, Nedjma Ousidhoum

*Warmth* (W) (often further broken down into*Trust* (T) and *Sociability* (S)) and *Competence* (C) are central dimensions along which people evaluate individuals and social groups (Fiske, 2018). While these constructs are well established in social psychology, they are only starting to get attention in NLP research through word-level lexicons, which do not fully capture their contextual expression in larger text units and discourse. In this work, we introduce*Warmth and Competence Sentences (W C-Sent)*, the first sentence-level dataset annotated for warmth and competence. The dataset includes over 1,600 English sentence–target pairs annotated along three dimensions: *trust* and *sociability* (components of *warmth*), and *competence*. The sentences in W C-Sent are social media posts that express attitudes and opinions about specific individuals or social groups (the targets of our annotations). We describe the data collection, annotation, and quality-control procedures in detail, and evaluate a range of large language models (LLMs) on their ability to identify trust, sociability, and competence in text. W C-Sent provides a new resource for analyzing warmth and competence in language and supports future research at the intersection of NLP and computational social science.

Chiwei Zhu, Benfeng Xu, Mingxuan Du, Shaohan Wang, Xiaorui Wang, Zhendong Mao, Yongdong Zhang

Deep research is emerging as a representative long-horizon task for large language model (LLM) agents. However, long trajectories in deep research often exceed model context limits, compressing token budgets for both evidence collection and report writing, and preventing effective test-time scaling. We introduce FS-Researcher, a file-system-based, dual-agent framework that scales deep research beyond the context window via a persistent workspace. Specifically, a Context Builder agent acts as a librarian which browses the internet, writes structured notes, and archives raw sources into a hierarchical knowledge base that can grow far beyond context length. A Report Writer agent then composes the final report section by section, treating the knowledge base as the source of facts. In this framework, the file system serves as a durable external memory and a shared coordination medium across agents and sessions, enabling iterative refinement beyond the context window. Experiments on two open-ended benchmarks (DeepResearch Bench and DeepConsult) show that FS-Researcher achieves state-of-the-art report quality across different backbone models. Further analyses demonstrate a positive correlation between final report quality and the computation allocated to the Context Builder, validating effective test-time scaling under the file-system paradigm. The code and data are open-sourced at https://github.com/Ignoramus0817/FS-Researcher.

Yifan Ji, Zhipeng Xu, Zhenghao Liu, Zulong Chen, Qian Zhang, Zhibo Yang, Junyang Lin, Yu Gu, Ge Yu, Maosong Sun

Key Information Extraction (KIE) from real-world documents remains challenging due to substantial variations in layout structures, visual quality, and task-specific information requirements. Recent Large Multimodal Models (LMMs) have shown promising potential for performing end-to-end KIE directly from document images. To enable a comprehensive and systematic evaluation across realistic and diverse application scenarios, we introduce UNIKIE-BENCH, a unified benchmark designed to rigorously evaluate the KIE capabilities of LMMs. UNIKIE-BENCH consists of two complementary tracks: a constrained-category KIE track with scenario-predefined schemas that reflect practical application needs, and an open-category KIE track that extracts any key information that is explicitly present in the document. Experiments on 15 state-of-the-art LMMs reveal substantial performance degradation under diverse schema definitions, long-tail key fields, and complex layouts, along with pronounced performance disparities across different document types and scenarios. These findings underscore persistent challenges in grounding accuracy and layout-aware reasoning for LMM-based KIE. All codes and datasets are available at https://github.com/NEUIR/UNIKIE-BENCH.

Hao Mi, Qiang Sheng, Shaofei Wang, Beizhe Hu, Yifan Sun, Zhengjia Wang, Hengqi Zeng, Yang Li, Danding Wang, Juan Cao

Large Language Models (LLMs) are prone to factual hallucinations, risking their reliability in real-world applications. Existing hallucination detectors mainly extract micro-level intrinsic patterns for uncertainty quantification or elicit macro-level self-judgments through verbalized prompts. However, these methods address only a single facet of the hallucination, focusing either on implicit neural uncertainty or explicit symbolic reasoning, thereby treating these inherently coupled behaviors in isolation and failing to exploit their interdependence for a holistic view. In this paper, we propose LaaB (Logical Consistency-as-a-Bridge), a framework that bridges neural features and symbolic judgments for hallucination detection. LaaB introduces a "meta-judgment" process to map symbolic labels back into the feature space. By leveraging the inherent logical bridge where response and meta-judgment labels are either the same or opposite based on the self-judgment’s semantics, LaaB aligns and integrates dual-view signals via mutual learning and enhances the hallucination detection. Extensive experiments on 4 public datasets, across 4 LLMs, against 8 baselines demonstrate the superiority of LaaB.

Truong Dinh Do, Nguyen-Khang Le, Le-Minh Nguyen

Speculative decoding accelerates large language model (LLM) inference through a draft-and-verify paradigm, yet existing methods face three key limitations: reliance on fixed draft templates that ignore device-specific verification costs, lack of mechanisms to assess draft token quality, and suboptimal tree expansion strategies. We introduce UniSpec, a training-free, lossless speculative decoding framework that enables robust, plug-and-play LLM acceleration across diverse hardware configurations and languages. UniSpec incorporates three novel components: (1) a device-aware calibration mechanism that determines the optimal draft size by measuring the acceptance-time trade-off on each target device; (2) a confidence score estimation module that assigns quality scores to n-grams based on the verifier’s token probabilities, enabling selective retention of high-quality draft candidates; and (3) an improved tree expansion strategy that broadens first-level exploration and applies threshold-based filtering to prune low-confidence nodes. To comprehensively evaluate multilingual performance, we create a comprehensive benchmark, covering seven languages across seven generation tasks. Experiments with various LLM architectures, hardware environments, and languages demonstrate that UniSpec consistently outperforms existing training-free methods, achieving speedups of up to 2.6x while maintaining output quality identical to standard autoregressive decoding. Our code and benchmark are publicly available.

Junhoo Lee, Seungyeon Kim, Nojun Kwak

Diffusion Language Models (DLMs) have emerged as a promising alternative to Autoregressive Language Models, yet their inference strategies largely rely on prefix-based prompting inherited from the autoregressive paradigm. In this paper, we propose Template Infilling (TI), a conditioning methodology tailored for DLMs. Unlike conventional prefix prompting, TI distributes structural anchors across the target response, establishing a global template before infilling masked segments. This enables structured conditioning that leverages the bidirectional generation process of DLMs. We evaluate TI on diverse benchmarks, including mathematical reasoning, code generation, and trip planning, achieving consistent improvements of 9.40%p over baseline prompting strategies. Furthermore, TI naturally supports multi-token generation settings, providing practical speed advantages while maintaining generation quality and robustness. Overall, our results highlight a DLM-specific conditioning paradigm for structured generation, suggesting a promising direction for inference methods tailored to diffusion-based language models.

Ling Shi, Xinwei Wu, Xiaohu Zhao, Hao Wang, Heng Liu, Yangyang Liu, Linlong Xu, Longyue Wang, Deyi Xiong, Weihua Luo

While mechanistic interpretability tools like Sparse Autoencoders (SAEs) can uncover meaningful features within Large Language Models (LLMs), a critical gap remains in transforming these insights into practical actions for model optimization. We bridge this gap with the hypothesis that data selection guided by a model’s internal task features is a effective training strategy. Inspired by this, we propose Interpretability-Guided Data Selection (IGDS), a framework that first identifies these causal task features through frequency recall and interventional filtering, then selects “Feature-Resonant Data” that maximally activates task features for fine-tuning. We validate IGDS on mathematical reasoning, summarization, and translation tasks within Gemma-2, LLaMA-3.1, and Qwen3 models. Our experiments demonstrate exceptional data efficiency: on the Math task, IGDS surpasses full-dataset fine-tuning by a remarkable **17.4%** on Gemma-2-2B while using only 50% of the data, and outperforms established baselines focused on data quality and diversity. Analysis confirms a strong positive correlation between feature amplification and task performance improvement. IGDS thus provides a direct and effective framework to enhance LLMs by leveraging their internal mechanisms, validating our core hypothesis.

Yee Man Choi, Xuehang Guo, Yi R. Fung, Qingyun Wang

Large Language Models (LLMs) have emerged as powerful assistants for scientific writing. However, concerns remain about the quality and reliability of the generated text, including citation accuracy and faithfulness. While most recent work relies on methods such as LLM-as-a-Judge, the reliability of LLM-as-a-Judge alone is also in doubt. In this work, we reframe citation evaluation as a problem of citation attribution alignment, which assesses whether LLM-generated citations match those a human author would include for the same text. We propose CiteGuard, a retrieval-aware agent framework designed to provide more faithful grounding for citation validation. CiteGuard improves over the prior baseline by 10 percentage points and achieves up to 68.1% accuracy on the CiteME benchmark, approaching human performance (69.2%). It also identifies alternative valid citations and demonstrates generalization ability for cross-domain citation attribution.

Zhiyuan Yao, Zishan Xu, Yifu Guo, Zhiguang Han, Cheng Yang, Shuo Zhang, Weinan Zhang, Xingshan Zeng, Weiwen Liu

With the rise of the Agent Web and Model Context Protocol (MCP), the agent ecosystem is evolving into an open collaborative network, exponentially increasing accessible tools. However, current architectures face severe scalability and generality bottlenecks. To address this, we propose ACE-Router, a pipeline for training history-aware routers to empower precise navigation in large-scale ecosystems. By leveraging a dependency-rich candidate Graph to synthesize multi-turn trajectories, we effectively train routers with dynamic context understanding to create the plug-and-play Light Routing Agent. Experiments on the real-world benchmarks MCP-Universe and MCP-Mark demonstrate superior performance. Notably, ACE-Router exhibits critical properties for the future Agent Web: it not only generalizes to multi-agent collaboration with minimal adaptation but also maintains exceptional robustness against noise and scales effectively to massive candidate spaces. These findings provide a strong empirical foundation for universal orchestration in open-ended ecosystems.Our code is available at https://github.com/euyis1019/ACE-Router.

Qingqing Lyu, Linjuan Wu, Yongliang Shen, Hengwei Liu, Hao Li, Shengpei Jiang, Yin Zhang, Weiming Lu

Despite the rapid progress of LLMs, their evaluation remains hindered by static, manually curated benchmarks with limited task coverage and poor adaptability to emerging domains. Existing automated approaches typically operate within fixed task schemas and often fail to autonomously discover new evaluation dimensions, limiting both scalability and effectiveness. To address these gaps, we propose AutoTaskEval, an automated framework that constructs domain-specific benchmarks directly from unstructured corpora. Using a refined Bloom’s Taxonomy, the framework systematically discovers tasks, enriches contextual grounding via iterative Socratic prompting, and generates diverse, progressively challenging evaluation instances. Applied to the complex and knowledge-intensive legal domain, AutoTaskEval uncovers a broader and more fine-grained task space than expert-curated benchmarks while producing high-quality instances that preserve established model-level evaluation trends. We further validate its robustness in a low-structure e-commerce review domain. Together, these results show that AutoTaskEval enables scalable, adaptive, and high-fidelity LLM assessment across domains and model families, advancing autonomous and capability-sensitive evaluation.

Yueqi Xie, Tao Qi, Jingwei Yi, Xiyuan Yang, Ryan Whalen, Junming Huang, Qian Ding, Yu Xie, Xing Xie, Fangzhao Wu

With the growing prevalence of generative AI, an increasing amount of content is no longer exclusively generated by humans but by generative AI models with human guidance. This shift presents notable challenges for the delineation of originality due to the varying degrees of human contribution in AI-assisted works. This study raises the research question of measuring human contribution in AI-assisted content generation and introduces a framework to address this question that is grounded in information theory. By calculating mutual information between human input and AI-assisted output relative to self-information of AI-assisted output, we quantify the proportional information contribution of humans in content generation. Our experimental results demonstrate that the proposed measure effectively discriminates between varying degrees of human contribution across multiple creative domains. To further enhance real-world applicability, we extend the framework to estimate the minimal necessary human contribution for any text without requiring human input and validate its effectiveness. We hope that this work lays a foundation for measuring human contributions in AI-assisted content generation in the era of generative AI.

Quyu Kong, Xu Zhang, Zhenyu Yang, Nolan Gao, Chen Liu, Panrong Tong, Chenglin Cai, Hanzhang Zhou, Jianan Zhang, Liangyu Chen 等

While AndroidWorld has become the dominant mobile-use benchmark due to its reproducible environment and deterministic evaluation, recent agents achieving over 90% success rates indicate saturation and motivate the need for greater challenge. In addition, its environment lacks key application categories, such as e-commerce and enterprise communication, and does not reflect realistic mobile-use scenarios characterized by vague user instructions and hybrid tool usage. We introduce MobileWorld, a substantially more challenging benchmark with 201 tasks across 20 applications that reflects real-world usage through long-horizon, cross-application workflows requiring nearly twice as many steps (27.8 vs. 14.3) and featuring significantly more multi-app tasks (62.2% vs. 9.5%) than AndroidWorld. MobileWorld balances production-grade utility and reproducible evaluation using open-source alternatives to industry standards (e.g., Mattermost for Slack), enabling full observability through source code modification and direct database access. Beyond standard GUI manipulation, MobileWorld introduces novel task categories including agent-user interaction and Model Context Protocol (MCP)-augmented tasks for evaluating agents in user-aware, hybrid-tool scenarios. We develop a planner-executor framework with extended action spaces supporting user interactions and MCP calls. Results show a sharp performance drop from AndroidWorld, with the best agentic framework and end-to-end model achieving 51.7% and 20.9% success rates, respectively, highlighting substantial room for future research.

Zhengxuan Lu, Dongfang Li, Yukun Shi, Beilun Wang, Longyue Wang, Baotian Hu

Current approaches to memory in Large Language Models (LLMs) predominantly rely on static Retrieval-Augmented Generation (RAG), which often results in scattered retrieval and fails to capture the structural dependencies required for complex reasoning. For autonomous agents, these passive and flat architectures lack the cognitive organization necessary to model the dynamic and associative nature of long-term interaction. To address this, we propose **S**tructured **E**pisodic **E**vent **M**emory (**SEEM**), a hierarchical framework that synergizes a graph memory layer for relational facts with a dynamic episodic memory layer for narrative progression. Grounded in cognitive frame theory, SEEM transforms interaction streams into structured Episodic Event Frames (EEFs) anchored by precise provenance pointers. Furthermore, we introduce an agentic associative fusion and Reverse Provenance Expansion (RPE) mechanism to reconstruct coherent narrative contexts from fragmented evidence. Experimental results on the LoCoMo and LongMemEval benchmarks demonstrate that SEEM significantly outperforms baselines, enabling agents to maintain superior narrative coherence and logical consistency.

Yang Ji, Ying Sun

Structured pruning offers a hardware-friendly approach for efficient LLM inference. Early static methods determine fixed subnetworks through offline calibration, suffering from performance degradation and calibration sensitivity. Recent methods explore input-adaptive pruning by selecting a subset of tokens as probes to estimate hidden activations for online pruning decisions.However, existing probe selection strategies fail to identify outlier-triggering tokens, and uniform layer-wise sparsity misaligns with heterogeneous outlier distributions, leading to critical channels being incorrectly pruned. Therefore, we propose OCP (Outlier-Centric Probing for structured pruning), a principled framework that prioritizes capturing outlier-triggering tokens rather than reconstructing full hidden distributions. Specifically, OCP includes three key components: (1) sensitivity-weighted probing for FFN layers that identifies outlier patterns via precomputed weight aggregations, (2) attention-accumulated probing that leverages preceding attention matrices to identify salient tokens, and (3) online adaptive sparsity allocation that dynamically adjusts layer-wise pruning based on history-guided outlier statistics. Extensive experiments on LLaMA2, LLaMA3, and OPT demonstrate that OCP consistently outperforms state-of-the-art methods across benchmarks, achieving up to 25% perplexity reduction at 1.6\times speedup.

Yuhang Niu, Hongyuan Xu, Ciyi Liu, Bofan Wei, Jiaqi Ye, Yanlong Wen, Xiaojie Yuan

Taxonomy Completion aims to automatically integrate new concepts into existing hierarchies. However, existing text-only methods suffer from a ”Sensory Gap”: they struggle to differentiate ambiguous definitions (e.g., Latte vs. Cappuccino) and miss visual grouping signals. Consequently, they often misinterpret lexical overlaps as hierarchical dependencies, leading to erroneous structural predictions. To bridge this, we propose VITC, a framework leveraging Visual Injection for Taxonomy Completion. By mapping synthesized images into intrinsic pseudo-tokens, we enable the text encoder to perform holistic structural reasoning. To address injection challenges, we introduce Adaptive Residual Fusion, which decouples magnitude from selection to prevent visual signals from being drowned out, and the Multimodal Guided Adaptive Reweighting strategy, which leverages cross-modal consensus (Mutual Rescue and Complementary Mining) to filter noise and identify hard negatives. Experiments on three datasets demonstrate that VITC achieves state-of-the-art performance, delivering an average absolute gain of over 19% in Hit@1. Code is available at https://github.com/nyh-a/VITC.

Qian Zhang, Xinye Li, Xiaokai Wu, Junhao Xu, Zhanyue Qin, Qingbin Liu, Junxian Cai, Xi Chen, Bolin Zhang, Zhiying Tu 等

Model Editing, also known as knowledge editing, is receiving increasing attention in the field of Large Language Models (LLMs). However, existing model editing approaches predominantly focus on knowledge-level or static visual domains, overlooking dynamic semantics. This paper exploratively applies six representative model editing methods (FT, IKE, MEND, SERAC, MEMIT and AlphaEdit) to Video Large Language Models (Vid-LLMs) and introduces the first benchmark specifically designed for Vid-LLMs editing—VMEB (Vid-LLMs Model Editing Benchmark)—systematically extending model editing research from static modalities to dynamic video scenarios. We position this work as a forward-looking benchmark and a foundational diagnostic study: in the video paradigm, our evaluation dimensions encompass traditional metrics including Reliability, Locality, and Generality, while also introducing a video-specific metric: Robustness. Based on experimental results, we analyze the strengths and limitations of existing model editing approaches, and identify new challenges and research directions for the future development of the model editing field within the context of multimodal and video paradigms. Our benchmark is available at https://github.com/Sakabamrisa/VMEB.