论文检索

Arkadiusz Modzelewski, Paweł Golik, Anna Kołos, Giovanni Da San Martino

Large Language Models (LLMs) can generate highly persuasive text, raising concerns about their misuse for propaganda, manipulation, and other harmful purposes. This leads us to our central question: Is LLM-generated persuasion more difficult to automatically detect than human-written persuasion? To address this, we categorize controllable generation approaches for producing persuasive content with LLMs and introduce Persuaficial, a high-quality multilingual benchmark covering six languages: English, German, Polish, Italian, French and Russian. Using this benchmark, we conduct extensive empirical evaluations comparing human-authored and LLM-generated persuasive texts. We find that although overtly persuasive LLM-generated texts can be easier to detect than human-written ones, subtle LLM-generated persuasion consistently degrades automatic detection performance. Beyond detection performance, we provide the first comprehensive linguistic analysis contrasting human and LLM-generated persuasive texts, offering insights that may guide the development of more interpretable and robust detection tools.

Zhezheng Hao, Hong Wang, Jian Luo, Jianqing Zhang, Yuyan Zhou, Qiang Lin, Can Wang, Hande Dong, Jiawei Chen

Large Language Model (LLM) agents are reshaping the industrial landscape. However, most practical agents remain human-designed because tasks differ widely, making them labor-intensive to build. This situation poses a central question: can we automatically create and adapt domain agents in the wild? While several recent approaches have sought to automate agent creation, they typically treat agent generation as a black-box procedure and rely solely on final performance metrics to guide the process. Such strategies overlook critical evidence explaining why an agent succeeds or fails, and often require high computational costs. To address these limitations, we propose ReCreate, an experience-driven framework for the automatic creation of domain agents. ReCreate systematically leverages agent interaction histories, which provide rich concrete signals on both the causes of success or failure and the avenues for improvement. Specifically, we introduce an agent-as-optimizer paradigm that effectively learns from experience via three key components: (i) an experience storage and retrieval mechanism for on-demand inspection; (ii) a reasoning–creating synergy pipeline that map execution experience into scaffold edits; and (iii) hierarchical updates that abstract instance-level details into reusable domain patterns. In experiments across diverse domains, ReCreate consistently outperforms human-designed agents and existing automated agent generation methods, even when starting from minimal seed scaffolds.

Xiangyu Wu, Yuwei Hu, Tianyu Cui, Yueying Tian, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang, Yang Yang, Jianfeng Lu

The path to fully autonomous web agents is currently hindered by a critical bottleneck: their limited ability to handle CAPTCHA. Existing agent benchmarks largely ignore this practical challenge, failing to evaluate an agent’s real-world capacity to solve CAPTCHA. To bridge this gap, we conduct a comprehensive analysis of real-world CAPTCHA distributions and introduce MirrorCAPTCHA, a benchmark annotated with Weighted Pass Rate and a newly proposed metric Completion Degree. MirrorCAPTCHA is designed to serve as a “mirror” that faithfully reflects the automation capabilities of agents in real scenarios. We filter 2095 websites from Common Crawl, identify the CAPTCHA deployed on these sites, and cluster them into 18 distinct categories using K-means algorithm. To ensure practicality, we extract a web subgraph from Common Crawl covering these websites and use random walks to simulate real-world CAPTCHA encounter frequencies, yielding a realistic measure of agents’ ability. Additionally, we develop a lightweight synthetic data pipeline to train Ovis2-Agent-CAPTCHA-8B, which significantly outperforms current state-of-the-art closed-source models on MirrorCAPTCHA, achieving a 9.4% higher average Weighted Pass Rate and a 2.13% higher average Completion Degree than the runner-up, Gemini-2.5-Pro.

Tobias Schreieder, Tim Schopf, Michael Färber

The increasing adoption of large language models (LLMs) has raised serious concerns about their reliability and trustworthiness. As a result, a growing body of research focuses on evidence-based text generation with LLMs, aiming to link model outputs to supporting evidence to ensure traceability and verifiability. However, the field is fragmented due to inconsistent terminology, isolated evaluation practices, and a lack of unified benchmarks. To bridge this gap, we systematically analyze 134 papers, introduce a unified taxonomy of evidence-based text generation with LLMs, and investigate 300 evaluation metrics across seven key dimensions. Thereby, we focus on approaches that use citations, attribution, or quotations for evidence-based text generation. Building on this, we examine the distinctive characteristics and representative methods in the field. Finally, we highlight open challenges and outline promising directions for future work.

Ahao Liu, Haitong Yang, Wang Chuanrong

Identical linguistic expressions can convey different sentiments across cultural contexts. Yet, current multilingual models often reduce language to mere symbolic representation, neglecting the cultural pragmatics that fundamentally shape affective semantics in Sentiment Analysis (SA). Due to this oversight, the systematic performance degradation of Small Multilingual Language Models (SMLMs) on culturally distant targets is frequently attributed to resource constraints. This perspective obscures the pivotal role of cultural pragmatics, an intrinsic determinant of affective semantics, and thereby conceals cultural misalignment as the principal structural bottleneck. In this paper, we conduct a comprehensive empirical study to quantify the influence of culture on cross-lingual sentiment transfer across 7 common SMLMs and 5 linguistically diverse languages. By fitting a Linear Mixed-Effects Model (LMEM) on over 300 experimental runs, we disentangle cultural factors from confounding variables. Our results reveal that Cultural Distance is a significant, independent, negative predictor of transfer performance. Furthermore, representation probing and qualitative error analysis uncover a pragmatic alignment paradox: while SMLMs encode cultural distinctions, they fail to map these representations to downstream sentiment labels in high-context cultures. Ultimately, our work enhances the interpretability of cross-lingual transfer failures by statistically isolating cultural misalignment as a structural barrier, distinct from the resource constraints typically blamed for poor performance.

Sirry Chen, Jieyi Wang, Wei Chen, Zhongyu Wei

Medical consultations are intrinsically speech-centric. However, most prior works focus on long-text-based interactions, which are cumbersome and patient-unfriendly. Recent advances in speech language models (SpeechLMs) have enabled more natural speech-based interaction, yet the scarcity of medical speech data and the inefficiency of directly fine-tuning on speech data jointly hinder the adoption of SpeechLMs in medical consultation. In this paper, we propose SpeechMedAssist, a SpeechLM natively capable of conducting speech-based multi-turn interactions with patients. By exploiting the architectural properties of SpeechLMs, we decouple the conventional one-stage training into a two-stage paradigm consisting of **(1) Knowledge Capability Injection via Text** and **(2) Modality Re-alignment with Limited Speech Data**, thereby reducing the requirement for medical speech data to only **10k** synthesized samples. To evaluate SpeechLMs for medical consultation scenarios, we design a benchmark comprising both single-turn question answering and multi-turn simulated interactions. Experimental results show that our model outperforms all baselines in both effectiveness and robustness in most evaluation settings.

Xiaoquan Yi, Haixing Wu, Haozhao Wang, Yichen Li, Yuhua Li, Rui Zhang, Ruixuan Li

The proliferation of Large Language Models (LLMs) has saturated social media platforms with hyper-realistic posts, rendering traditional detection methods that rely on low-level artifacts or unimodal statistics increasingly ineffective. In this work, we identify a fundamental semantic distinction: humans tend to complement visual content with additional context, while LLMs predominantly describe the visual information. To capture this, UMPIRE employs an orthogonal semantic decomposition mechanism that disentangles textual embeddings into redundant and complementary components. An adaptive gating module dynamically weighs these components to reflect diverse communicative styles. To enforce the desired geometric structure, we introduce a latent contrastive redundancy regularization loss that encourages LLM-generated content to exhibit high semantic redundancy, while human-written content emphasizes complementarity. Experimental results demonstrate that UMPIRE significantly outperforms state-of-the-art detection methods across multiple datasets, achieving up to a 5.38% improvement in accuracy.

Philipp Seeberger, Steffen Freisinger, Tobias Bocklet, Korbinian Riedhammer

Multimedia event extraction aims to jointly identify events and their arguments across multiple modalities, such as text and images, to support more comprehensive event understanding. While recent work reports steady and substantial progress, the reliability and comparability of these results critically depend on consistent and rigorous evaluation. In this work, we present the first systematic analysis of evaluation pitfalls in multimedia event extraction and identify three major sources of issues: inconsistent data processing, inconsistent task assumptions, and overly relaxed evaluation settings. We demonstrate, through a series of controlled experiments under a strict evaluation framework, that minor evaluation choices can cause large performance variations and lead to overestimation of a model’s ability to ground real-world events across modalities. Our findings highlight the need for comparable evaluation standards and encourage a shift toward more rigorous evaluation in multimedia event extraction.

Xianjing Han, Bin Zhu, Shiqi Hu, Franklin Mingzhe Li, Patrick Carrington, Roger Zimmermann, Jingjing Chen

Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text–video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object’s state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action–object interactions into regular, novel, and compositional scenarios to probe both in-distribution performance and generalization. We evaluate six representative open-source and proprietary T2V models using both human user study and multimodal large language model (MLLM)–based automatic evaluation. Our results show that, despite strong performance on semantic and scene alignment, current T2V models consistently struggle with accurate and temporally consistent object state changes, especially in novel and compositional settings. These findings position OSC as a key bottleneck in text-to-video generation and establish OSCBench as a diagnostic benchmark for advancing state-aware video generation models. Project page: https://hanxjing.github.io/OSCBench.

Ying He, Sihang Jiang, Xingzhou Chen, Zhouhong Gu, Yiwei Gu, Minggui HE, Shimin Tao, Mahongxia, Yanghua Xiao

Cultural taboo safety is essential for deploying large language models (LLMs), as culturally insensitive outputs may cause offense or even social harm. However, existing cultural benchmarks primarily assess cultural knowledge or values biases, while overlooking whether LLMs can recognize and respect cultural taboos, especially when taboos are implicitly hidden in seemingly harmless questions. Besides, cultural taboos are implicit, and context-dependent, thus poss unique challenges for reliable evaluation. To address these gaps, we introduce **CulShield**, the first public benchmark dedicated to evaluating and improving the cultural taboo safety of LLMs. CulShield spans 77 countries and regions, and includes over 2,020 taboos. It evaluates models along both explicit knowledge and implicit behaviors.Experiments on several advanced LLMs (e.g., GPT-4o-mini, Gemini-2.5-pro) reveal a clear "knowledge-behavior gap": models often fail to apply known taboos during interaction. We further show that variations in linguistic context can significantly affect LLMs’ cultural taboo safety. Code and data is accessible here: https://anonymous.4open.science/r/CulShield-7A0E.

Songxin Qu, Tai-Ping Sun, Yun-Jie Wang, Huan-Yu Liu, Cheng Xue, Xiao-Fan Xu, Han Fang, Yang Yang, Yu-Chun Wu, Guo-Ping Guo 等

Large language models (LLMs) show strong capabilities in general reasoning but typically lack reliability in scientific domains like quantum mechanics, which demand strict adherence to physical constraints. This limitation arises from the scarcity of verifiable training resources and the inadequacy of coarse feedback signals in standard alignment paradigms. To address the data challenge, we introduce QuantumQA, a large-scale dataset constructed via a task-adaptive strategy and a hybrid verification protocol that combines deterministic solvers with semantic auditing to guarantee scientific rigor. Building on this foundation, we propose the verification-aware reward model (VRM) tailored for Reinforcement Learning with Verifiable Rewards (RLVR), which employs an adaptive reward fusion (ARF) mechanism to dynamically integrate deterministic signals from a scientific execution suite (SES) with multidimensional semantic evaluations for precise supervision. Experimental results demonstrate that our method consistently outperforms baselines and general-purpose preference models. Notably, our optimized 8B model achieves performance competitive with proprietary models, validating that incorporating verifiable, rule-based feedback into the reinforcement learning loop offers a parameter-efficient alternative to pure scaling.

Jihao Gu, Qihang Ai, Yingyao Wang, Pi Bu, Jingxuan Xing, Yue Cao, Zekun Zhu, Wei Jiang, Ziming Wang, Yingxiu Zhao 等

Vision-language model-based mobile agents have gained the ability to understand complex instructions and mobile screenshots, benefiting from reinforcement learning paradigms like Group Relative Policy Optimization (GRPO). However, existing approaches centers on offline training or local action-level rewards often trap agents in local optima, hindering effective exploration and error correction with the environment. Crucially, we find that directly applying task-level rewards often leads to convergence difficulties due to the sparse nature of GUI interactions. To address these challenges, we present Mobile-R1, a systematic training recipe that bridges atomic action execution and strategic task completion. We propose a hierarchical curriculum consisting of three stages: (1) format alignment for reasoning structure, (2) on-policy exploration with verifiable action feedback to ground basic execution, and (3) multi-turn task-level training with realistic environment to unlock exploration and self-correction. This hierarchical strategy effectively bootstraps the agent, significantly enhancing its capability for exploration and self-correction (the “Eureka” moments). Furthermore, addressing the critical scarcity of diverse GUI data in non-English ecosystems, we contribute a comprehensive Chinese mobile dataset covering 28 applications with 24,521 high-quality manual annotations, and establish a rigorous benchmark with 500 trajectories. We will open source all resources, including the dataset, benchmark, model weight, and codes: https://mobile-r1.github.io/Mobile-R1/.

Arya Shah, Deepali Mishra, Chaklam Silpasuwanchai

Large language models increasingly serve as conversational agents that adopt personas and role-play characters at user request. This capability, while valuable, raises concerns about sycophancy: the tendency to provide responses that validate users rather than prioritize factual accuracy. While prior work has established that sycophancy poses risks to AI safety and alignment, the relationship between specific personality traits of adopted personas and the degree of sycophantic behavior remains unexplored. We present a systematic investigation of how persona agreeableness influences sycophancy across 13 small, open-weight language models ranging from 0.6B to 20B parameters. We develop a benchmark comprising 275 personas evaluated on NEO-IPIP agreeableness subscales and expose each persona to 4,950 sycophancy-eliciting prompts spanning 33 topic categories. Our analysis reveals that 9 of 13 models exhibit statistically significant positive correlations between persona agreeableness and sycophancy rates, with Pearson correlations reaching r = 0.87 and effect sizes as large as Cohen’s d = 2.33. These findings demonstrate that agreeableness functions as a reliable predictor of persona-induced sycophancy, with direct implications for the deployment of role-playing AI systems and the development of alignment strategies that account for personality-mediated deceptive behaviors

Yixuan Nan, Xixun Lin, Yanmin Shang, Ge Zhang, Zheng Fang, Fang Fang, Yanan Cao

Entity alignment (EA) aims to identify entities across different knowledge graphs (KGs) that refer to the same real-world object and plays a critical role in knowledge fusion and integration. Traditional EA methods mainly rely on knowledge representation learning, but their performance is often limited under noisy or sparsely supervised scenarios. Recently, large language models (LLMs) have been introduced to EA and achieved notable improvements by leveraging rich semantic knowledge. However, existing LLM-based EA approaches typically treat LLMs as black-box decision makers, resulting in limited interpretability, and the direct use of large-scale triples substantially increases inference cost. To address these challenges, we propose EA-Agent, a reasoning-driven agent for EA. EA-Agent formulates EA as a structured reasoning process with multi-step planning and execution, enabling interpretable alignment decisions. Within this process, it introduces attribute and relation triple selectors to filter redundant triples before feeding them into the LLM, effectively addressing efficiency challenges. Experimental results on three benchmark datasets demonstrate that EA-Agent consistently outperforms existing EA methods and achieves state-of-the-art performance. The source code is available at https://anonymous.4open.science/r/EA-Agent-5696.

Janvijay Singh, Dilek Hakkani-Tür

Large language models solve complex tasks by generating long reasoning chains, achieving higher accuracy at the cost of increased computational cost and reduced ability to isolate functionally relevant reasoning. Prior work on compact reasoning shortens such chains through probabilistic sampling, heuristics, or supervision from frontier models, but offers limited insight into whether models internally encode token-level functional importance for answer generation. We address this gap diagnostically and propose greedy pruning, a likelihood-preserving deletion procedure that iteratively removes reasoning tokens whose removal minimally degrades model likelihood under a specified objective, yielding length-controlled reasoning chains. We evaluate pruned reasoning in a distillation framework and show that students trained on pruned chains outperform a frontier-model–supervised compression baseline at matched reasoning lengths. Finally, our analysis reveals systematic pruning patterns and shows that attention scores can predict greedy pruning ranks, further suggesting that models encode a nontrivial functional importance structure over reasoning tokens.

Zhanyu Liu, Qingguo Hu, Ante Wang, Chenqing Liu, Zhishang Xiang, Hui Li, Delai Qiu, Jinsong Su

Reinforcement Learning with Verifiable Reward (RLVR) has proven effective for training reasoning-oriented large language models, but existing methods largely assume high-resource settings with abundant training data. In low-resource scenarios, RLVR is prone to more severe entropy collapse, which substantially limits exploration and degrades reasoning performance. To address this issue, we propose **H**ybrid-domain **E**ntropy dynamics **AL**ignment (HEAL), a framework tailored for few-shot RLVR. HEAL first selectively incorporates high-value general-domain data to promote more diverse exploration. Then, we introduce Entropy Dynamics Alignment (EDA), a reward mechanism that aligns trajectory-level entropy dynamics between the target and general domains, capturing both entropy magnitude and fine-grained variation. Through this alignment, EDA not only further mitigates entropy collapse but also encourages the policy to acquire more diverse exploration behaviors from the general domain. Experiments across multiple domains show that HEAL consistently improves few-shot RLVR performance. Notably, using only 32 target-domain samples, HEAL matches or even surpasses full-shot RLVR trained with 1K target-domain samples.

Xiaoyue Lu, Xianglin Yang, Haijun Liu, Jiahao Liu, Kuntai Cai, Yan Xiao, Jin Song Dong

The widespread integration of Large Language Models (LLMs) necessitates rigorous and systematic safety evaluation. Existing paradigms either rely on constructed benchmarks to assess safety from predefined perspectives, or employ dynamic red-teaming to probe potential vulnerabilities. While effective, these approaches face challenges, as they depend heavily on expert domain knowledge, offer limited systematic guarantees, and are vulnerable to rapid obsolescence. To address these limitations, we introduce a novel framework POLARIS that brings the rigor of specification-based software testing to AI safety. POLARIS first compiles unstructured natural-language policies into First-Order Logic (FOL) representations, establishing a traceable link between high-level rules and concrete test cases. This formalization enables the construction of a Semantic Policy Graph, where complex policy violation scenarios are encoded as traversable paths. By systematically exploring this graph, POLARIS uncovers compositional violation patterns, which are then instantiated into executable natural-language test queries, enabling coverage-driven and reproducible safety testing. Experiments demonstrate that POLARIS achieves higher policy coverage and attack success counts compared to established baselines. Crucially, by bridging formal methods and AI safety, POLARIS provides a principled, automated approach to ensuring LLMs adhere to safety-critical policies with verifiable traceability.

Eric Onyame, Akash Ghosh, Subhadip Baidya, Sriparna Saha, Xiuying Chen, Chirag Agarwal

While large language models (LLMs) have shown to perform well on monolingual mathematical and commonsense reasoning, they remain unreliable for multilingual medical reasoning applications, hindering their deployment in multilingual healthcare settings. We address this by first introducing CURE-Med-Bench, a high-quality multilingual medical reasoning dataset with open-ended reasoning queries with a single verifiable answer, spanning thirteen languages, including underrepresented languages such as Amharic, Yoruba, and Swahili. Building on this dataset, we propose CURE-Med, a curriculum-informed reinforcement learning framework that integrates code-switching-aware supervised fine-tuning and Group Relative Policy Optimization to jointly improve logical correctness and language stability. Across thirteen languages, our approach consistently outperforms strong baselines and scales effectively, achieving 85.21% language consistency and 54.35% logical correctness at 7B parameters and 94.96% language consistency and 70.04% logical correctness at 32B parameters. These results support reliable and equitable multilingual medical reasoning in LLMs. The code and dataset will be made publicly available upon acceptance.

Michal Štefánik, Timothee Mickus, Marek Kadlčík, Bertram Højer, Michal Spiegel, Raúl Vázquez, Aman Sinha, Josef Kuchař, Philipp Mondorf, Pontus Stenetorp

Prior work has shown that large language models (LLMs) often converge to accurate input embedding for numbers, based on sinusoidal representations.In this work, we demonstrate that these representations are in fact strikingly systematic, to the point of being almost perfectly universal: different LLM families develop equivalent sinusoidal structures, and number representations are broadly interchangeable in a large swathe of experimental setups.We show that properly factoring in this characteristic is crucial when it comes to assessing how accurately LLMs encode numeric and other ordinal information, and that mechanistically enhancing this sinusoidality can also lead to reductions of LLMs’ arithmetic errors.

Bojun Jin, Jianzhu Bao, Yang Sun, Yice Zhang, Ruifeng Xu

Argument generation is a fundamental NLP task that aims to automatically produce persuasive arguments.Effective human argumentation is inherently complex and multifaceted, integrating argumentative strategies, appropriate styles, and adaptation to target audiences, etc.However, existing studies focus on limited control signals such as topic, stance, or key aspects, failing to capture this complexity.As LLMs advance, the lack of benchmarks evaluating multifaceted argumentative control becomes a critical bottleneck.To address this, we introduce ArgGenBench, a novel benchmark containing complex instructions that integrate multi-dimensional control, including topic, stance, length, style, strategy, audience, and key points.Extensive evaluation across 15 LLMs reveals significant limitations: even the best-performing model achieves only 42.7% win rate against human-verified references.These results highlight the challenge of controlled argument generation and establish ArgGenBench as a rigorous testbed for developing more capable systems.