论文检索

Shouju Wang, Fenglin Yu, Xirui Liu, Xiaoting Qin, Jue Zhang, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan

The increasing autonomy of LLM agents in handling sensitive communications, accelerated by Model Context Protocol (MCP) and Agent-to-Agent (A2A) frameworks, creates urgent privacy challenges. While recent work reveals significant gaps between LLMs’ privacy Q&A performance and their agent behavior, existing benchmarks remain limited to static, simplified scenarios. We present PrivacyChecker, a model-agnostic, contextual integrity based mitigation approach that effectively reduces privacy leakage from 36.08% to 7.30% on DeepSeek-R1 and from 33.06% to 8.32% on GPT-4o, all while preserving task helpfulness. We also introduce PrivacyLens-Live, transforming static benchmarks into dynamic MCP and A2A environments that reveal substantially higher privacy risks in practical. Our modular mitigation approach integrates seamlessly into agent protocols through three deployment strategies, providing practical privacy protection for the emerging agentic ecosystem. Our data and code will be made available at https://aka.ms/privacy_in_action.

Hanqing Li, Sharika Mahadevan, Kiran Jyothi Sheena, Henry Liang, Diego Klabjan

We propose a new, training-free method, Graph Reasoning via Retrieval Augmented Framework (GRRAF), that harnesses retrieval-augmented generation (RAG) alongside the code-generation capabilities of large language models (LLMs) to address a wide range of graph reasoning tasks. In GRRAF, the target graph is stored in a graph database, and the LLM is prompted to generate executable code queries that retrieve the necessary information. This approach circumvents the limitations of existing methods that require extensive finetuning or depend on predefined algorithms, and it incorporates an error feedback loop with a time-out mechanism to ensure both correctness and efficiency. Experimental evaluations on the GraphInstruct dataset reveal that GRRAF achieves 100% accuracy on most graph reasoning tasks, including cycle detection, bipartite graph checks, shortest path computation, and maximum flow, while maintaining consistent token costs regardless of graph sizes. Imperfect but still very high performance is observed on subgraph matching. Notably, GRRAF scales effectively to large graphs with up to 10,000 nodes.

Bo Peng, Zhiheng Wang, Heyang Gong, Chaochao Lu

In modern dialogue systems, the ability to implicitly infer user backgrounds from conversations and leverage this information for personalized assistance is crucial. However, the scarcity of high-quality data remains a fundamental challenge to evaluating and improving this capability. Traditional dataset construction methods are labor-intensive, resource-demanding, and raise privacy concerns. To address these issues, we propose a novel approach for automatic synthetic data generation and introduce the **I**mplicit **P**ersonalized **Dialog**ue (**IP-Dialog**) benchmark along with a training dataset, covering 10 tasks and 12 user attribute types. Additionally, we develop a systematic evaluation framework with four metrics to assess both attribute awareness and reasoning capabilities. We further propose five causal graphs to elucidate models’ reasoning pathways during implicit personalization. Extensive experiments yield insightful observations and prove the reliability of our dataset.

Elias Bassani, Ignacio Sanchez

The risk of generative AI systems providing unsafe information has raised significant concerns, emphasizing the need for safety guardrails. To mitigate this risk, guardrail models are increasingly used to detect unsafe content in human-AI interactions, complementing the safety alignment of Large Language Models. Despite recent efforts to evaluate those models’ effectiveness, their robustness to input mutations and adversarial attacks remains largely unexplored. In this paper, we present a comprehensive evaluation of 15 state-of-the-art guardrail models, assessing their robustness to: a) input mutations, such as typos, keywords camouflage, ciphers, and veiled expressions, and b) adversarial attacks designed to bypass models’ safety alignment. Those attacks exploit LLMs capabilities like instruction-following, role-playing, personification, reasoning, and coding, or introduce adversarial tokens to induce model misbehavior. Our results reveal that most guardrail models can be evaded with simple input mutations and are vulnerable to adversarial attacks. For instance, a single adversarial token can deceive them 44.5% of the time on average. The limitations of the current generation of guardrail models highlight the need for more robust safety guardrails.

Qinhong Lin, Zhongliang Yang, Yuang Cai, Dingfu Yu, Xuan Xu, Yu Li, Linna Zhou

Retrieval-Augmented Generation (RAG) plays a critical role in mitigating hallucinations and improving factual accuracy for Large Language Models (LLMs). While dynamic retrieval techniques aim to determine retrieval timing and content based on model intrinsic needs, existing approaches struggle to generalize effectively in black-box model scenarios. To address this limitation, we propose the Semantic Contribution-Aware Adaptive Retrieval (SCAAR) framework. SCAAR iteratively leverages the semantic importance of words in upcoming sentences to dynamically adjust retrieval thresholds and filter information, retaining the top-\alpha% most semantically significant words for constructing retrieval queries. We comprehensively evaluate SCAAR against baseline methods across four long-form, knowledge-intensive generation datasets using four models. Our method achieved the highest score on each dataset with GPT-4o. Extensive experiments also analyze the impact of various hyperparameters within the framework. Our results demonstrate SCAAR’s superior or competitive performance, showcasing its ability to effectively detect model retrieval needs and construct efficient retrieval queries for relevant knowledge about problem-solving in black-box scenarios. Our code is available on https://github.com/linqinhong/SAC.

Zhihui Yang, Yupei Wang, Kaijie Mo, Zhe Zhao, Renfen Hu

Despite significant progress in multimodal language models (LMs), it remains unclear whether visual grounding enhances their understanding of embodied knowledge compared to text-only models. To address this question, we propose a novel embodied knowledge understanding benchmark based on the perceptual theory from psychology, encompassing visual, auditory, tactile, gustatory, olfactory external senses, and interoception. The benchmark assesses the models’ perceptual abilities across different sensory modalities through vector comparison and question-answering tasks with over 1,700 questions. By comparing 30 state-of-the-art LMs, we surprisingly find that vision-language models (VLMs) do not outperform text-only models in either task. Moreover, the models perform significantly worse in the visual dimension compared to other sensory dimensions. Further analysis reveals that the vector representations are easily influenced by word form and frequency, and the models struggle to answer questions involving spatial perception and reasoning. Our findings underscore the need for more effective integration of embodied knowledge in LMs to enhance their understanding of the physical world.

Valentin Barriere, Nahuel Gomez, Léo Hemamou, Sofia Callejas, Brian Ravenet

Aiming towards improving current computational models of humor detection, we propose a new multimodal dataset of stand-up comedies, in seven languages: English, French, Spanish, Italian, Portuguese, Hungarian and Czech. Our dataset of more than 330 hours %, is at the time of writing the biggest available for this type of task, and the most diverse. The whole dataset is automatically annotated in laughter (from the audience), and the subpart left for model validation is manually annotated.% Contrary to contemporary approaches, we do not frame the task of humor detection as a binary sequence classification, but as word-level sequence labeling, in order to take into account all the context of the sequence and to capture the continuous joke tagging mechanism typically occurring in natural conversations. As par with unimodal baselines results, we propose a method for e propose a method to enhance the automatic laughter detection based on Audio Speech Recognition errors. Our code and data are available online: https://tinyurl.com/EMNLPHumourStandUpAnonym

Patara Trirat, Jae-Gil Lee

The growing use of smartphones and IoT devices necessitates efficient time-series analysis on resource-constrained hardware, which is critical for sensing applications such as human activity recognition and air quality prediction. Recent efforts in hardware-aware neural architecture search (NAS) automate architecture discovery for specific platforms; however, none focus on general time-series analysis with edge deployment. Leveraging the problem-solving and reasoning capabilities of large language models (LLM), we propose ***MONAQ***, a novel framework that reformulates NAS into ***M***ulti-***O***bjective ***N***eural ***A***rchitecture ***Q***uerying tasks. *MONAQ* is equipped with *multimodal query generation* for processing multimodal time-series inputs and hardware constraints, alongside an *LLM agent-based multi-objective search* to achieve deployment-ready models via code generation. By integrating numerical data, time-series images, and textual descriptions, *MONAQ* improves an LLM’s understanding of time-series data. Experiments on fifteen datasets demonstrate that *MONAQ*-discovered models outperform both handcrafted models and NAS baselines while being more efficient.

Yifan Zhu, Chao Zhang, Xin Shi, Xueqiao Zhang, Yi Yang, Yawei Luo

Large Language Models (LLMs)-based Multi-Agent Systems (MAS) exhibit remarkable problem-solving and task planning capabilities across diverse domains due to their specialized agentic roles and collaborative interactions. However, this also amplifies the severity of security risks under MAS attacks. To address this, we introduce \textbf{MASTER}, a novel security research framework for MAS, focusing on diverse \textbf{R}ole configurations and \textbf{T}opological structures across various scenarios. MASTER offers an automated construction process for different MAS setups and an information-flow-based interaction paradigm. To tackle MAS security challenges in varied scenarios, we design a scenario-adaptive, extensible attack strategy utilizing role and topological information, which dynamically allocates targeted, domain-specific attack tasks for collaborative agent execution. Our experiments demonstrate that such an attack, leveraging role and topological information, exhibits significant destructive potential across most models. Additionally, we propose corresponding defense strategies, substantially enhancing MAS resilience across diverse scenarios. We anticipate that our framework and findings will provide valuable insights for future research into MAS security challenges.

Yijiang River Dong, Tiancheng Hu, Yinhong Liu, Ahmet Üstün, Nigel Collier

While Reinforcement Learning from Human Feedback (RLHF) is widely used to align Large Language Models (LLMs) with human preferences, it typically assumes homogeneous preferences across users, overlooking diverse human values and minority viewpoints.Although personalized preference learning addresses this by tailoring separate preferences for individual users, the field lacks standardized methods to assess its effectiveness. We present a multi-faceted evaluation framework that measures not only performance but also fairness, unintended effects, and adaptability across varying levels of preference divergence. Through extensive experiments comparing eight personalization methods across three preference datasets, we demonstrate that performance differences between methods could reach 36% when users strongly disagree, and personalization can introduce up to 20% safety misalignment. These findings highlight the critical need for holistic evaluation approaches to advance the development of more effective and inclusive preference learning systems.

Ikuya Yamada, Ryokan Ri, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo

Dense retrievers often struggle with queries involving less-frequent entities due to their limited entity knowledge. We propose the Knowledgeable Passage Retriever (KPR), a BERT-based retriever enhanced with a context-entity attention layer and dynamically updatable entity embeddings. This design enables KPR to incorporate external entity knowledge without retraining. Experiments on three datasets demonstrate that KPR consistently improves retrieval accuracy, with particularly large gains on the EntityQuestions dataset. When built on the off-the-shelf bge-base retriever, KPR achieves state-of-the-art performance among similarly sized models on two datasets. Models and code are released at https://github.com/knowledgeable-embedding/knowledgeable-embedding.

Runchao Li, Yao Fu, Mu Sheng, Xianxuan Long, Haotian Yu, Pan Li

The efficacy of Large Language Models (LLMs) in long-context tasks is often hampered by the substantial memory footprint and computational demands of the Key-Value (KV) cache. Current compression strategies, including token eviction and learned projections, frequently lead to biased representations—either by overemphasizing recent/high-attention tokens or by repeatedly degrading information from earlier context—and may require costly model retraining. We present FAEDKV (Frequency-Adaptive Infinite-Window for KV cache), a novel, training-free KV cache compression framework that ensures unbiased information retention. FAEDKV operates by transforming the KV cache into the frequency domain using a proposed Infinite-Window Fourier Transform (IWDFT). This approach allows for the equalized contribution of all tokens to the compressed representation, effectively preserving both early and recent contextual information. A preliminary frequency ablation study identifies critical spectral components for layer-wise, targeted compression. Experiments on LongBench benchmark demonstrate FAEDKV’s superiority over existing methods by up to 22%. In addition, our method shows superior, position-agnostic retrieval accuracy on the Needle-In-A-Haystack task compared to compression based approaches.

Abhay Gupta, Jacob Cheung, Philip Meng, Shayan Sayyed, Kevin Zhu, Austen Liao, Sean O’Brien

The diversity of human language, shaped by social, cultural, and regional influences, presents significant challenges for natural language processing (NLP) systems. Existing benchmarks often overlook intra-language variations, leaving speakers of non-standard dialects underserved. To address this gap, we introduce EnDive (English Diversity), a benchmark that evaluates seven state-of-the-art (SOTA) large language models (LLMs) across tasks in language understanding, algorithmic reasoning, mathematics, and logic. Our framework translates Standard American English datasets into five underrepresented dialects using few-shot prompting with verified examples from native speakers, and compares these translations against rule-based methods via fluency assessments, preference tests, and semantic similarity metrics. Human evaluations confirm high translation quality, with average scores of at least 6.02/7 for faithfulness, fluency, and formality. By filtering out near-identical translations, we create a challenging dataset that reveals significant performance disparities—models consistently underperform on dialectal inputs compared to Standard American English (SAE). EnDive thus advances dialect-aware NLP by uncovering model biases and promoting more equitable language technologies.

Jingen Qu, Lijun Li, Bo Zhang, Yichen Yan, Jing Shao

Multimodal large language models (MLLMs) are rapidly evolving, presenting increasingly complex safety challenges. However, current dataset construction methods, which are risk-oriented, fail to cover the growing complexity of real-world multimodal safety scenarios (RMS). And due to the lack of a unified evaluation metric, their overall effectiveness remains unproven. This paper introduces a novel image-oriented self-adaptive dataset construction method for RMS, which starts with images and end constructing paired text and guidance responses. Using the image-oriented method, we automatically generate an RMS dataset comprising 35,610 image–text pairs with guidance responses. Additionally, we introduce a standardized safety dataset evaluation metric: fine-tuning a safety judge model and evaluating its capabilities on other safety datasets. Extensive experiments on various tasks demonstrate the effectiveness of the proposed image-oriented pipeline. The results confirm the scalability and effectiveness of the image-oriented approach, offering a new perspective for the construction of real-world multimodal safety datasets.

Juli Bakagianni, Korbinian Randl, Guido Rocchietti, Cosimo Rulli, Franco Maria Nardini, Salvatore Trani, Aron Henriksson, Anna Romanova, John Pavlopoulos

Food safety demands timely detection, regulation, and public communication, yet the lack of structured datasets hinders Natural Language Processing (NLP) research. We present and release a new dataset of human-written and Large Language Model (LLM)-generated summaries of food safety documents, plus food safety related metadata. We evaluate its utility on three NLP tasks directly reflecting food safety practices: multilabel classification for organizing documents into domain-specific categories; document retrieval for accessing regulatory and scientific evidence; and question answering via retrieval-augmented generation that improves factual accuracy.We show that LLM summaries perform comparably or better than human ones across tasks. We also demonstrate clustering of summaries for event tracking and compliance monitoring. This dataset enables NLP applications that support core food safety practices, including the organization of regulatory and scientific evidence, monitoring of compliance issues, and communication of risks to the public.

Ramya Keerthy Thatikonda, Wray Buntine, Ehsan Shareghi

The recent successful paradigm of solving logical reasoning problems with tool-augmented large language models (LLMs) leverages translation of natural language (NL) statements into First-Order Logic (FOL) and external theorem provers. However, the correctness of FOL statements, comprising operators and text, often go unverified due to the lack of a reliable evaluation metric for comparing generated and ground-truth FOLs. In this paper, we conduct a comprehensive study on the sensitivity of existing metrics—NL, FOL, and graph-based— and their alignment with LLM as a judge on FOL evaluation to measure robustness. We introduce operator and text-based perturbations to ground-truth FOL statements to assess metric sensitivity. We then evaluate metric robustness by comparing them against LLMs judgement. Our empirical findings highlight a clear oversensitivity in the n-gram metric BLEU for text perturbations. The operator perturbation affects the semantic graph metric Smatch++ for structural changes, and the FOL metric for specific operator changes. We observe a closer alignment between BertScore and LLM judgement, proving the importance of semantic evaluation. Additionally, we show that combining metrics enhances both robustness and sensitivity compared to using individual metrics.

Xu Huang, Wenhao Zhu, Hanxu Hu, Conghui He, Lei Li, Shujian Huang, Fei Yuan

Existing multilingual benchmarks focus primarily on language understanding tasks. There is a lack of benchmarks to measure comprehensive critical capabilities of large language models (LLMs) across diverse languages, including instruction following, reasoning, code generation, and long context understanding. To bridge this gap, we develop BenchMAX, a multi-way multilingual benchmark that covers 10 diverse tasks, to evaluate LLMs’ general abilities across many languages. To ensure high data quality, each sample is post-edited by three native annotators after machine-translating from English into 16 languages. Extensive experiments on BenchMAX reveal uneven utilization of core capabilities across languages, emphasizing the performance gaps that scaling model size alone does not resolve. BenchMAX serves as a comprehensive multilingual evaluation platform, providing a promising test bed to promote the development of multilingual language models. The dataset and code are publicly accessible.

Yitao Long, Tiansheng Hu, Yilun Zhao, Arman Cohan, Chen Zhao

Large Language Models (LLMs) frequently hallucinate to long-form questions, producing plausible yet factually incorrect answers. A common mitigation strategy is to provide attribution to LLM outputs. However, existing benchmarks primarily focus on simple attribution that retrieves supporting textual evidence as references. We argue that in real-world scenarios such as financial applications, attribution goes beyond reference retrieval.We introduce FinLFQA, a benchmark designed to evaluate the ability of LLMs to generate long-form answers to complex financial questions with reliable and nuanced attributions. FinLFQA evaluates three critical aspects of attribution through human annotations: (1) supporting evidence extracted from financial reports, (2) intermediate numerical reasoning steps, and (3) domain-specific financial knowledge that informs the reasoning process.We further provide an automatic evaluation framework covering both answer quality and attribution quality. Through extensive experiments on eight LLMs across multiple attribution-generation paradigms, we find that fine-grained metrics are important to distinguish model capabilities, that end-to-end generation achieves comparable performance to post-hoc approaches, and that iterative refinement only helps when guided by external feedback.

Qian Xiong, Yuekai Huang, Ziyou Jiang, Zhiyuan Chang, Yujia Zheng, Tianhao Li, Mingyang Li

The emergence of the tool agent paradigm has broadened the capability boundaries of the Large Language Model (LLM), enabling it to complete more complex tasks. However, the effectiveness of this paradigm is limited due to the issue of parameter failure during its execution. To explore this phenomenon and propose corresponding suggestions, we first construct a parameter failure taxonomy in this paper. We derive five failure categories from the invocation chain of a mainstream tool agent. Then, we explore the correlation between three different input sources and failure categories by applying 15 input perturbation methods to the input. Experimental results show that parameter name hallucination failure primarily stems from inherent LLM limitations, while issues with input sources mainly cause other failure patterns. To improve the reliability and effectiveness of tool-agent interactions, we propose corresponding improvement suggestions, including standardizing tool return formats, improving error feedback mechanisms, and ensuring parameter consistency.

Jiaxin Guo, Daimeng Wei, Yuanchang Luo, Hengchao Shang, Zongyao Li, Jinlong Yang, Zhanglin Wu, Zhiqiang Rao, Shimin Tao, Hao Yang

With the widespread application of Large Language Models (LLMs) in the field of Natural Language Processing (NLP), enhancing their performance has become a research hotspot. This paper presents a novel multi-prompt ensemble decoding approach designed to bolster the generation quality of LLMs by leveraging the aggregation of outcomes from multiple prompts. Given a unique input X, we submit n variations of prompts with X to LLMs in batch mode to decode and derive probability distributions. For each token prediction, we calculate the ensemble probability by averaging the n probability distributions within the batch, utilizing this aggregated probability to generate the token. This technique is dubbed Inner-Batch Ensemble. To facilitate efficient batch inference, we implement a Left-Padding strategy to maintain uniform input lengths across the n prompts. Through extensive experimentation on diverse NLP tasks, including code generation, text simplification and machine translation, we demonstrate the efficacy of our method in enhancing LLM performance. The results show substantial improvements in pass@k rates, LENS metrics and BLEU scores over conventional methods.