Zimu Wang

Xiaozhi Wang, Yulin Chen, Ning Ding et al. · tencent-ai, tsinghua

The diverse relationships among real-world events, including coreference, temporal, causal, and subevent relations, are fundamental to understanding natural languages. However, two drawbacks of existing datasets limit event relation extraction (ERE) tasks: (1) Small scale. Due to the annotation complexity, the data scale of existing datasets is limited, which cannot well train and evaluate data-hungry models. (2) Absence of unified annotation. Different types of event relations naturally interact with each other, but existing datasets only cover limited relation types at once, which prevents models from taking full advantage of relation interactions. To address these issues, we construct a unified large-scale human-annotated ERE dataset MAVEN-ERE with improved annotation schemes. It contains 103,193 event coreference chains, 1,216,217 temporal relations, 57,992 causal relations, and 15,841 subevent relations, which is larger than existing datasets of all the ERE tasks by at least an order of magnitude. Experiments show that ERE on MAVEN-ERE is quite challenging, and considering relation interactions with joint learning can improve performances. The dataset and source codes can be obtained from https://github.com/THU-KEG/MAVEN-ERE.

Hao Peng, Xiaozhi Wang, Feng Yao et al. · tsinghua

Event understanding aims at understanding the content and relationship of events within texts, which covers multiple complicated information extraction tasks: event detection, event argument extraction, and event relation extraction. To facilitate related research and application, we present an event understanding toolkit OmniEvent, which features three desiderata: (1) Comprehensive. OmniEvent supports mainstream modeling paradigms of all the event understanding tasks and the processing of 15 widely-used English and Chinese datasets. (2) Fair. OmniEvent carefully handles the inconspicuous evaluation pitfalls reported in Peng et al. (2023), which ensures fair comparisons between different models. (3) Easy-to-use. OmniEvent is designed to be easily used by users with varying needs. We provide off-the-shelf models that can be directly deployed as web services. The modular framework also enables users to easily implement and evaluate new event understanding models with OmniEvent. The toolkit (https://github.com/THU-KEG/OmniEvent) is publicly released along with the demonstration website and video (https://omnievent.xlore.cn/).

Hao Peng, Xiaozhi Wang, Jianhui Chen et al. · tsinghua

In-context learning (ICL) has become the default method for using large language models (LLMs), making the exploration of its limitations and understanding the underlying causes crucial. In this paper, we find that ICL falls short of handling specification-heavy tasks, which are tasks with complicated and extensive task specifications, requiring several hours for ordinary humans to master, such as traditional information extraction tasks. The performance of ICL on these tasks mostly cannot reach half of the state-of-the-art results. To explore the reasons behind this failure, we conduct comprehensive experiments on 18 specification-heavy tasks with various LLMs and identify three primary reasons: inability to specifically understand context, misalignment in task schema comprehension with humans, and inadequate long-text understanding ability. Furthermore, we demonstrate that through fine-tuning, LLMs can achieve decent performance on these tasks, indicating that the failure of ICL is not an inherent flaw of LLMs, but rather a drawback of existing alignment methods that renders LLMs incapable of handling complicated specification-heavy tasks via ICL. To substantiate this, we perform dedicated instruction tuning on LLMs for these tasks and observe a notable improvement. We hope the analyses in this paper could facilitate advancements in alignment methods enabling LLMs to meet more sophisticated human demands.

Hongbin Na, Zimu Wang, Zhaoming Chen et al.

We present an overview of PsyDefDetect, the shared task on detecting levels of psychological defense mechanisms in emotional support dialogues, co-located with BioNLP@ACL 2026. Grounded in the clinically validated Defense Mechanism Rating Scales (DMRS) framework, the task asks systems to classify a target seeker utterance, given its preceding dialogue context, into one of nine categories: seven hierarchical DMRS levels plus two auxiliary labels. Participants worked on PsyDefConv, a newly released corpus of 200 dialogues and 2336 help-seeker utterances annotated under DMRS with substantial inter-annotator agreement. The task attracted 172 participants on CodaBench who produced 563 submissions, with 21 teams officially registering their results for the final ranking. The best system achieved a macro F1-score of 0.420, surpassing the strongest fine-tuned baseline reported in the dataset paper by a notable margin, yet leaving clear headroom. Our analysis highlights (i) a persistent tendency to over-predict the majority High-Adaptive class, (ii) a widening gap between accuracy and macro-F1 that reveals class-imbalance sensitivity, and (iii) the value of theory-aware and LLM-based approaches for fine-grained defensive-function classification. We release all task materials and invite the community to continue work on this novel intersection of clinical psychology and NLP.

Zimu Wang, Wei Wang, Qi Chen et al.

Deep learning-based natural language processing (NLP) models, particularly pre-trained language models (PLMs), have been revealed to be vulnerable to adversarial attacks. However, the adversarial examples generated by many mainstream word-level adversarial attack models are neither valid nor natural, leading to the loss of semantic maintenance, grammaticality, and human imperceptibility. Based on the exceptional capacity of language understanding and generation of large language models (LLMs), we propose LLM-Attack, which aims at generating both valid and natural adversarial examples with LLMs. The method consists of two stages: word importance ranking (which searches for the most vulnerable words) and word synonym replacement (which substitutes them with their synonyms obtained from LLMs). Experimental results on the Movie Review (MR), IMDB, and Yelp Review Polarity datasets against the baseline adversarial attack models illustrate the effectiveness of LLM-Attack, and it outperforms the baselines in human and GPT-4 evaluation by a significant margin. The model can generate adversarial examples that are typically valid and natural, with the preservation of semantic meaning, grammaticality, and human imperceptibility.

Stephanie Fong, Zimu Wang, Guilherme C. Oliveira et al.

The medical adoption of NLP tools requires interpretability by end users, yet traditional explainable AI (XAI) methods are misaligned with clinical reasoning and lack clinician input. We introduce CHiRPE (Clinical High-Risk Prediction with Explainability), an NLP pipeline that takes transcribed semi-structured clinical interviews to: (i) predict psychosis risk; and (ii) generate novel SHAP explanation formats co-developed with clinicians. Trained on 944 semi-structured interview transcripts across 24 international clinics of the AMP-SCZ study, the CHiRPE pipeline integrates symptom-domain mapping, LLM summarisation, and BERT classification. CHiRPE achieved over 90% accuracy across three BERT variants and outperformed baseline models. Explanation formats were evaluated by 28 clinical experts who indicated a strong preference for our novel concept-guided explanations, especially hybrid graph-and-text summary formats. CHiRPE demonstrates that clinically-guided model development produces both accurate and interpretable results. Our next step is focused on real-world testing across our 24 international sites.

Jiayuan Ma, Hongbin Na, Zimu Wang et al.

Mental manipulation severely undermines mental wellness by covertly and negatively distorting decision-making. While there is an increasing interest in mental health care within the natural language processing community, progress in tackling manipulation remains limited due to the complexity of detecting subtle, covert tactics in conversations. In this paper, we propose Intent-Aware Prompting (IAP), a novel approach for detecting mental manipulations using large language models (LLMs), providing a deeper understanding of manipulative tactics by capturing the underlying intents of participants. Experimental results on the MentalManip dataset demonstrate superior effectiveness of IAP against other advanced prompting strategies. Notably, our approach substantially reduces false negatives, helping detect more instances of mental manipulation with minimal misjudgment of positive cases. The code of this paper is available at https://github.com/Anton-Jiayuan-MA/Manip-IAP.

Dongzhe Fan, Chuanhao Ji, Zimu Wang et al.

Graph-based retrieval-augmented generation (GraphRAG) has recently emerged as a powerful paradigm for knowledge-intensive question answering, especially for tasks that require structured evidence organization and multi-hop reasoning. However, existing GraphRAG systems are typically built in a one-size-fits-all manner, relying on a fixed retrieval framework and a single, often large and costly, generator LLM for all queries. This static design limits their ability to adapt to the complexity of varying questions and often incurs unnecessary computational cost. To fill in the gap, we propose GraphRAG-Router, a cost-efficient framework that adopts a hierarchical routing strategy to coordinate heterogeneous GraphRAGs and generator LLMs. Specifically, GraphRAG-Router is first warmed up through supervised fine-tuning and then optimized with a two-stage reinforcement learning procedure, whose second stage introduces a curriculum cost-aware reward to encourage difficulty-aware and economical generator allocation. Extensive experiments on six general-domain and multi-hop QA benchmarks show that GraphRAG-Router consistently outperforms state-of-the-art baselines, reducing the overuse of large LLMs by nearly 30% while maintaining strong generalization capability.

Zimu Wang, Yuling Shi, Mengfan Li et al.

Code efficiency is a fundamental aspect of software quality, yet how to harness large language models (LLMs) to optimize programs remains challenging. Prior approaches have sought for one-shot rewriting, retrieved exemplars, or prompt-based search, but they do not explicitly distill reusable optimization knowledge, which limits generalization beyond individual instances. In this paper, we present EffiSkill, a framework for code-efficiency optimization that builds a portable optimization toolbox for LLM-based agents. The key idea is to model recurring slow-to-fast transformations as reusable agent skills that capture both concrete transformation mechanisms and higher-level optimization strategies. EffiSkill adopts a two-stage design: Stage I mines Operator and Meta Skills from large-scale slow/fast program pairs to build a skill library; Stage II applies this library to unseen programs through execution-free diagnosis, skill retrieval, plan composition, and candidate generation, without runtime feedback. Results on EffiBench-X show that EffiSkill achieves higher optimization success rates, improving over the strongest baseline by 3.69 to 12.52 percentage points across model and language settings. These findings suggest that mechanism-level skill reuse provides a useful foundation for execution-free code optimization, and that the resulting skill library can serve as a reusable resource for broader agent workflows.

Yaling Shen, Stephanie Fong, Yiwen Jiang et al.

The increasing integration of large language models (LLMs) into mental health applications necessitates robust frameworks for evaluating professional safety alignment. Current evaluative approaches primarily rely on refusal-based safety signals, which offer limited insight into the nuanced behaviors required in clinical practice. In mental health, clinically inadequate refusals can be perceived as unempathetic and discourage help-seeking. To address this gap, we move beyond refusal-centric metrics and introduce \texttt{PsychEthicsBench}, the first principle-grounded benchmark based on Australian psychology and psychiatry guidelines, designed to evaluate LLMs' ethical knowledge and behavioral responses through multiple-choice and open-ended tasks with fine-grained ethicality annotations. Empirical results across 14 models reveal that refusal rates are poor indicators of ethical behavior, revealing a significant divergence between safety triggers and clinical appropriateness. Notably, we find that domain-specific fine-tuning can degrade ethical robustness, as several specialized models underperform their base backbones in ethical alignment. PsychEthicsBench provides a foundation for systematic, jurisdiction-aware evaluation of LLMs in mental health, encouraging more responsible development in this domain.

Hongbin Na, Zimu Wang, Zhaoming Chen et al.

Psychological defenses are strategies, often automatic, that people use to manage distress. Rigid or overuse of defenses is negatively linked to mental health and shapes what speakers disclose and how they accept or resist help. However, defenses are complex and difficult to reliably measure, particularly in clinical dialogues. We introduce PsyDefConv, a dialogue corpus with help seeker utterances labeled for defense level, and DMRS Co-Pilot, a four-stage pipeline that provides evidence-based pre-annotations. The corpus contains 200 dialogues and 4709 utterances, including 2336 help seeker turns, with labeling and Cohen's kappa 0.639. In a counterbalanced study, the co-pilot reduced average annotation time by 22.4%. In expert review, it averaged 4.62 for evidence, 4.44 for clinical plausibility, and 4.40 for insight on a seven-point scale. Benchmarks with strong language models in zero-shot and fine-tuning settings demonstrate clear headroom, with the best macro F1-score around 30% and a tendency to overpredict mature defenses. Corpus analyses confirm that mature defenses are most common and reveal emotion-specific deviations. We will release the corpus, annotations, code, and prompts to support research on defensive functioning in language.

Xueqing Peng, Zhuohan Xie, Yupeng Cao et al.

As AI agents improve, the central question is no longer whether they can solve isolated well-defined financial tasks, but whether they can reliably carry out financial professional work. Existing financial benchmarks offer only a partial view of this ability, as they primarily evaluate static competencies such as question answering, retrieval, summarization, and classification. We introduce Herculean, the first skilled benchmark for agentic financial intelligence spanning four representative workflows, including Trading, Hedging, Market Insights, and Auditing. Each workflow is instantiated as a standardized MCP-based skill environment with its own tools, interaction dynamics, constraints, and success criteria, enabling consistent end-to-end assessment of heterogeneous agent systems. Across frontier agents, we find agents perform relatively well on Trading and Market Insights, but struggle substantially on Hedging and Auditing, where long-horizon coordination, state consistency, and structured verification are critical. Overall, our results point to a key gap in current agents in turning financial reasoning into dependable workflow execution in high-stakes financial workflows.

Tong Chen, Zimu Wang, Yiyi Miao et al.

Medical fact-checking has become increasingly critical as more individuals seek medical information online. However, existing datasets predominantly focus on human-generated content, leaving the verification of content generated by large language models (LLMs) relatively unexplored. To address this gap, we introduce MedFact, the first evidence-based Chinese medical fact-checking dataset of LLM-generated medical content. It consists of 1,321 questions and 7,409 claims, mirroring the complexities of real-world medical scenarios. We conduct comprehensive experiments in both in-context learning (ICL) and fine-tuning settings, showcasing the capability and challenges of current LLMs on this task, accompanied by an in-depth error analysis to point out key directions for future research. Our dataset is publicly available at https://github.com/AshleyChenNLP/MedFact.

Muyi Bao, Changyu Zeng, Yifan Wang et al.

Transformer-based deep neural networks have achieved remarkable success across various computer vision tasks, largely attributed to their long-range self-attention mechanism and scalability. However, most transformer architectures embed images into uniform, grid-based vision tokens, neglecting the underlying semantic meanings of image regions, resulting in suboptimal feature representations. To address this issue, we propose Fuzzy Token Clustering Transformer (FTCFormer), which incorporates a novel clustering-based downsampling module to dynamically generate vision tokens based on the semantic meanings instead of spatial positions. It allocates fewer tokens to less informative regions and more to represent semantically important regions, regardless of their spatial adjacency or shape irregularity. To further enhance feature extraction and representation, we propose a Density Peak Clustering-Fuzzy K-Nearest Neighbor (DPC-FKNN) mechanism for clustering center determination, a Spatial Connectivity Score (SCS) for token assignment, and a channel-wise merging (Cmerge) strategy for token merging. Extensive experiments on 32 datasets across diverse domains validate the effectiveness of FTCFormer on image classification, showing consistent improvements over the TCFormer baseline, achieving gains of improving 1.43% on five fine-grained datasets, 1.09% on six natural image datasets, 0.97% on three medical datasets and 0.55% on four remote sensing datasets. The code is available at: https://github.com/BaoBao0926/FTCFormer/tree/main.

Changyu Zeng, Yifan Wang, Zimu Wang et al.

Recent advancements in 2D multimodal large language models (MLLMs) have significantly improved performance in vision-language tasks. However, extending these capabilities to 3D environments remains a distinct challenge due to the complexity of spatial reasoning. Nevertheless, existing 3D benchmarks often lack fine-grained numerical reasoning task annotations, limiting MLLMs' ability to perform precise spatial measurements and complex numerical reasoning. To address this gap, we introduce NUMINA, the first Natural Understanding benchmark for Multi-dimensional Intelligence and Numerical reasoning Abilities to enhance multimodal indoor perceptual understanding. NUMINA features multi-scale annotations and various question-answer pairs, generated using NUMINA-Flow, an automated annotation pipeline that integrates LLM rewriting and rule-based self-verification. We evaluate the performance of various state-of-the-art LLMs on NUMINA following the Chat-Scene framework, demonstrating that current LLMs struggle with multimodal numerical reasoning, particularly in performing precise computations such as distance and volume estimation, highlighting the need for further advancements in 3D models. The dataset and source codes can be obtained from https://github.com/fengshun124/NUMINA.

Xiaoqiang Kang, Zimu Wang, Xiaobo Jin et al.

Solving tabular math word problems (TMWPs) has become a critical role in evaluating the mathematical reasoning ability of large language models (LLMs), where large-scale TMWP samples are commonly required for LLM fine-tuning. Since the collection of high-quality TMWP datasets is costly and time-consuming, recent research has concentrated on automatic TMWP generation. However, current generated samples usually suffer from issues of either correctness or diversity. In this paper, we propose a Template-driven LLM-paraphrased (TeLL) framework for generating high-quality TMWP samples with diverse backgrounds and accurate tables, questions, answers, and solutions. To this end, we first extract templates from existing real samples to generate initial problems, ensuring correctness. Then, we adopt an LLM to extend templates and paraphrase problems, obtaining diverse TMWP samples. Furthermore, we find the reasoning annotation is important for solving TMWPs. Therefore, we propose to enrich each solution with illustrative reasoning steps. Through the proposed framework, we construct a high-quality dataset TabMWP-TeLL by adhering to the question types in the TabMWP dataset, and we conduct extensive experiments on a variety of LLMs to demonstrate the effectiveness of TabMWP-TeLL in improving TMWP solving performance. The code and data of this paper are available at: https://github.com/Jason8Kang/TELL.

Hongbin Na, Yining Hua, Zimu Wang et al.

Mental health is increasingly critical in contemporary healthcare, with psychotherapy demanding dynamic, context-sensitive interactions that traditional NLP methods struggle to capture. Large Language Models (LLMs) offer significant potential for addressing this gap due to their ability to handle extensive context and multi-turn reasoning. This review introduces a conceptual taxonomy dividing psychotherapy into interconnected stages--assessment, diagnosis, and treatment--to systematically examine LLM advancements and challenges. Our comprehensive analysis reveals imbalances in current research, such as a focus on common disorders, linguistic biases, fragmented methods, and limited theoretical integration. We identify critical challenges including capturing dynamic symptom fluctuations, overcoming linguistic and cultural biases, and ensuring diagnostic reliability. Highlighting future directions, we advocate for continuous multi-stage modeling, real-time adaptive systems grounded in psychological theory, and diversified research covering broader mental disorders and therapeutic approaches, aiming toward more holistic and clinically integrated psychotherapy LLMs systems.

Jianfei He, Lilin Wang, Jiaying Wang et al.

Identifying offensive language is essential for maintaining safety and sustainability in the social media era. Though large language models (LLMs) have demonstrated encouraging potential in social media analytics, they lack thorough evaluation when in offensive language detection, particularly in multilingual environments. We for the first time evaluate multilingual offensive language detection of LLMs in three languages: English, Spanish, and German with three LLMs, GPT-3.5, Flan-T5, and Mistral, in both monolingual and multilingual settings. We further examine the impact of different prompt languages and augmented translation data for the task in non-English contexts. Furthermore, we discuss the impact of the inherent bias in LLMs and the datasets in the mispredictions related to sensitive topics.

Hongbin Na, Zimu Wang, Mieradilijiang Maimaiti et al.

Large language models (LLMs) have demonstrated promising potential in various downstream tasks, including machine translation. However, prior work on LLM-based machine translation has mainly focused on better utilizing training data, demonstrations, or pre-defined and universal knowledge to improve performance, with a lack of consideration of decision-making like human translators. In this paper, we incorporate Thinker with the Drift-Diffusion Model (Thinker-DDM) to address this issue. We then redefine the Drift-Diffusion process to emulate human translators' dynamic decision-making under constrained resources. We conduct extensive experiments under the high-resource, low-resource, and commonsense translation settings using the WMT22 and CommonMT datasets, in which Thinker-DDM outperforms baselines in the first two scenarios. We also perform additional analysis and evaluation on commonsense translation to illustrate the high effectiveness and efficacy of the proposed method.

Xingxuan Zhang, Gang Ren, Han Yu et al.

We argue that progress toward general intelligence requires complementary foundation models grounded in language, the physical world, and structured data. This report presents LimiX-16M and LimiX-2M, two instantiations of our large structured-data models (LDMs). Both models treat structured data as a joint distribution over variables and missingness, thus capable of addressing a wide range of tabular tasks through query-based conditional prediction via a single model. They are pretrained using masked joint-distribution modeling with an episodic, context-conditional objective, supporting rapid, training-free adaptation at inference. We evaluate LimiX models across 11 large structured-data benchmarks with broad regimes of sample size, feature dimensionality, class number, categorical-to-numerical feature ratio, missingness, and sample-to-feature ratios. LimiX-16M consistently surpasses strong baselines, as shown in Figure 1 and Figure 2. The superiority holds across a wide range of tasks, such as classification, regression, missing value imputation, and data generation, often by substantial margins, while avoiding task-specific architectures or bespoke training per task. Notably, LimiX-2M delivers strong results under tight compute and memory budgets. We also present the first scaling law study for LDMs, revealing how data and model scaling jointly influence downstream performance and offering quantitative guidance for tabular foundation modeling. All LimiX models are publicly accessible under Apache 2.0.

Shuo Yan, Ruochen Li, Ziming Luo et al. · microsoft-research

Large language model (LLM) agents have demonstrated remarkable potential in advancing scientific discovery. However, their capability in the fundamental yet crucial task of reproducing code from research papers, especially in the NLP domain, remains underexplored. This task includes unique complex reasoning challenges in the intellectual synthesis of abstract concepts and the comprehension of code repositories with interdependent files. Motivated by this gap, we present LMR-BENCH, a benchmark designed to systematically evaluate the capability of LLM agents on code reproduction from Language Modeling Research. It consists of 28 code reproduction tasks derived from 23 research papers published in top-tier NLP venues over the past five years, spanning nine fundamental categories. Models are provided with a research paper, a code repository containing one or more masked functions, and instructions for implementing these functions. We conduct extensive experiments in standard prompting and LLM agent settings with state-of-the-art LLMs, evaluating the accuracy of unit tests and performing LLM-based evaluation of code correctness. Experimental results reveal that even the most advanced models still exhibit persistent limitations in scientific reasoning and code synthesis, highlighting critical gaps in LLM agents' ability to autonomously reproduce scientific research

Lu Qian, Yuqi Wang, Zimu Wang et al.

In domain-specific contexts, particularly mental health, abstractive summarization requires advanced techniques adept at handling specialized content to generate domain-relevant and faithful summaries. In response to this, we introduce a guided summarizer equipped with a dual-encoder and an adapted decoder that utilizes novel domain-specific guidance signals, i.e., mental health terminologies and contextually rich sentences from the source document, to enhance its capacity to align closely with the content and context of guidance, thereby generating a domain-relevant summary. Additionally, we present a post-editing correction model to rectify errors in the generated summary, thus enhancing its consistency with the original content in detail. Evaluation on the MentSum dataset reveals that our model outperforms existing baseline models in terms of both ROUGE and FactCC scores. Although the experiments are specifically designed for mental health posts, the methodology we've developed offers broad applicability, highlighting its versatility and effectiveness in producing high-quality domain-specific summaries.

Qingyang Xu, Yaling Shen, Stephanie Fong et al.

The increasing use of large language models (LLMs) in mental healthcare raises safety concerns in high-stakes therapeutic interactions. A key challenge is distinguishing therapeutic empathy from maladaptive validation, where supportive responses may inadvertently reinforce harmful beliefs or behaviors in multi-turn conversations. This risk is largely overlooked by existing red-teaming frameworks, which focus mainly on generic harms or optimization-based attacks. To address this gap, we introduce Personality-based Client Simulation Attack (PCSA), the first red-teaming framework that simulates clients in psychological counseling through coherent, persona-driven client dialogues to expose vulnerabilities in psychological safety alignment. Experiments on seven general and mental health-specialized LLMs show that PCSA substantially outperforms four competitive baselines. Perplexity analysis and human inspection further indicate that PCSA generates more natural and realistic dialogues. Our results reveal that current LLMs remain vulnerable to domain-specific adversarial tactics, providing unauthorized medical advice, reinforcing delusions, and implicitly encouraging risky actions.

Yiwen Jiang, Deval Mehta, Siyuan Yan et al.

Multimodal Large Language Models (MLLMs) have shown promise in visual-textual reasoning, with Multimodal Chain-of-Thought (MCoT) prompting significantly enhancing interpretability. However, existing MCoT methods rely on rationale-rich datasets and largely focus on inter-object reasoning, overlooking the intra-object understanding crucial for image classification. To address this gap, we propose WISE, a Weak-supervision-guided Step-by-step Explanation method that augments any image classification dataset with MCoTs by reformulating the concept-based representations from Concept Bottleneck Models (CBMs) into concise, interpretable reasoning chains under weak supervision. Experiments across ten datasets show that our generated MCoTs not only improve interpretability by 37% but also lead to gains in classification accuracy when used to fine-tune MLLMs. Our work bridges concept-based interpretability and generative MCoT reasoning, providing a generalizable framework for enhancing MLLMs in fine-grained visual understanding.

Haotan Guo, Jianfei He, Jiayuan Ma et al.

Phonetic Cloaking Replacement (PCR), defined as the deliberate use of homophonic or near-homophonic variants to hide toxic intent, has become a major obstacle to Chinese content moderation. While this problem is well-recognized, existing evaluations predominantly rely on rule-based, synthetic perturbations that ignore the creativity of real users. We organize PCR into a four-way surface-form taxonomy and compile \ours, a dataset of 500 naturally occurring, phonetically cloaked offensive posts gathered from the RedNote platform. Benchmarking state-of-the-art LLMs on this dataset exposes a serious weakness: the best model reaches only an F1-score of 0.672, and zero-shot chain-of-thought prompting pushes performance even lower. Guided by error analysis, we revisit a Pinyin-based prompting strategy that earlier studies judged ineffective and show that it recovers much of the lost accuracy. This study offers the first comprehensive taxonomy of Chinese PCR, a realistic benchmark that reveals current detectors' limits, and a lightweight mitigation technique that advances research on robust toxicity detection.

Xiangyu Zhao, Yaling Shen, Yiwen Jiang et al.

Depression is one of the most prevalent mental health disorders globally. In recent years, multi-modal data, such as speech, video, and transcripts, has been increasingly used to develop AI-assisted depression assessment systems. Large language models have further advanced this field due to their strong language understanding and generalization capabilities. However, conventional LLMs remain text-centric and cannot process the rich non-verbal cues found in audio and visual modalities, which are critical components in mental health evaluation. While multi-modal LLMs offer a promising direction, few are tailored for psychological applications. In this study, we propose a novel multi-modal LLM framework for depression detection. Our approach augments an audio language model with visual understanding and aligns audio-visual features at the timestamp level. This fine-grained alignment improves modeling of temporal dynamics across modalities while reducing the need for extensive training data and computational resources. Experiments on the DAIC-WoZ dataset demonstrate that our model outperforms both single-modality approaches and previous multi-modal methods. Moreover, the proposed framework can be extended to incorporate additional physiological signals, paving the way for broader clinical applications beyond mental health.

Tianshi Cai, Guanxu Li, Nijia Han et al.

We introduce FinDebate, a multi-agent framework for financial analysis, integrating collaborative debate with domain-specific Retrieval-Augmented Generation (RAG). Five specialized agents, covering earnings, market, sentiment, valuation, and risk, run in parallel to synthesize evidence into multi-dimensional insights. To mitigate overconfidence and improve reliability, we introduce a safe debate protocol that enables agents to challenge and refine initial conclusions while preserving coherent recommendations. Experimental results, based on both LLM-based and human evaluations, demonstrate the framework's efficacy in producing high-quality analysis with calibrated confidence levels and actionable investment strategies across multiple time horizons.

Xiaoqiang Kang, Shengen Wu, Zimu Wang et al.

Existing table understanding methods face challenges due to complex table structures and intricate logical reasoning. While supervised finetuning (SFT) dominates existing research, reinforcement learning (RL), such as Group Relative Policy Optimization (GRPO), has shown promise but struggled with low initial policy accuracy and coarse rewards in tabular contexts. In this paper, we introduce Table-R1, a three-stage RL framework that enhances multimodal table understanding through: (1) Warm-up that prompts initial perception and reasoning capabilities, (2) Perception Alignment GRPO (PA-GRPO), which employs continuous Tree-Edit-Distance Similarity (TEDS) rewards for recognizing table structures and contents, and (3) Hint-Completion GRPO (HC-GRPO), which utilizes fine-grained rewards of residual steps based on the hint-guided question. Extensive experiments demonstrate that Table-R1 can boost the model's table reasoning performance obviously on both held-in and held-out datasets, outperforming SFT and GRPO largely. Notably, Qwen2-VL-7B with Table-R1 surpasses larger specific table understanding models (e.g., Table-LLaVA 13B), even achieving comparable performance to the closed-source model GPT-4o on held-in datasets, demonstrating the efficacy of each stage of Table-R1 in overcoming initialization bottlenecks and reward sparsity, thereby advancing robust multimodal table understanding.

Fuyu Xing, Zimu Wang, Wei Wang et al.

The proliferation of multimedia content necessitates the development of effective Multimedia Event Extraction (M2E2) systems. Though Large Vision-Language Models (LVLMs) have shown strong cross-modal capabilities, their utility in the M2E2 task remains underexplored. In this paper, we present the first systematic evaluation of representative LVLMs, including DeepSeek-VL2 and the Qwen-VL series, on the M2E2 dataset. Our evaluations cover text-only, image-only, and cross-media subtasks, assessed under both few-shot prompting and fine-tuning settings. Our key findings highlight the following valuable insights: (1) Few-shot LVLMs perform notably better on visual tasks but struggle significantly with textual tasks; (2) Fine-tuning LVLMs with LoRA substantially enhances model performance; and (3) LVLMs exhibit strong synergy when combining modalities, achieving superior performance in cross-modal settings. We further provide a detailed error analysis to reveal persistent challenges in areas such as semantic precision, localization, and cross-modal grounding, which remain critical obstacles for advancing M2E2 capabilities.

Huixin Zhong, Yanan Liu, Qi Cao et al.

As large language models (LLMs) integrate into collaborative teams, their social conformity -- the tendency to align with majority opinions -- has emerged as a key concern. In humans, conformity arises from informational influence (rational use of group cues for accuracy) or normative influence (social pressure for approval), with uncertainty moderating this balance by shifting from purely analytical to heuristic processing. It remains unclear whether these human psychological mechanisms apply to LLMs. This study adapts the information cascade paradigm from behavioral economics to quantitatively disentangle the two drivers to investigate the moderate effect. We evaluated nine leading LLMs across three decision-making scenarios (medical, legal, investment), manipulating information uncertainty (q = 0.667, 0.55, and 0.70, respectively). Our results indicate that informational influence underpins the models' behavior across all contexts, with accuracy and confidence consistently rising with stronger evidence. However, this foundational mechanism is dramatically modulated by uncertainty. In low-to-medium uncertainty scenarios, this informational process is expressed as a conservative strategy, where LLMs systematically underweight all evidence sources. In contrast, high uncertainty triggers a critical shift: while still processing information, the models additionally exhibit a normative-like amplification, causing them to overweight public signals (beta > 1.55 vs. private beta = 0.81).

Zimu Wang, Jiashuo Liu, Hao Zou et al.

Massive amounts of data are the foundation of data-driven recommendation models. As an inherent nature of big data, data heterogeneity widely exists in real-world recommendation systems. It reflects the differences in the properties among sub-populations. Ignoring the heterogeneity in recommendation data could limit the performance of recommendation models, hurt the sub-populational robustness, and make the models misled by biases. However, data heterogeneity has not attracted substantial attention in the recommendation community. Therefore, it inspires us to adequately explore and exploit heterogeneity for solving the above problems and assisting data analysis. In this work, we focus on exploring two representative categories of heterogeneity in recommendation data that is the heterogeneity of prediction mechanism and covariate distribution and propose an algorithm that explores the heterogeneity through a bilevel clustering method. Furthermore, the uncovered heterogeneity is exploited for two purposes in recommendation scenarios which are prediction with multiple sub-models and supporting debias. Extensive experiments on real-world data validate the existence of heterogeneity in recommendation data and the effectiveness of exploring and exploiting data heterogeneity in recommendation.

Yue He, Zimu Wang, Peng Cui et al.

In spite of the tremendous development of recommender system owing to the progressive capability of machine learning recently, the current recommender system is still vulnerable to the distribution shift of users and items in realistic scenarios, leading to the sharp decline of performance in testing environments. It is even more severe in many common applications where only the implicit feedback from sparse data is available. Hence, it is crucial to promote the performance stability of recommendation method in different environments. In this work, we first make a thorough analysis of implicit recommendation problem from the viewpoint of out-of-distribution (OOD) generalization. Then under the guidance of our theoretical analysis, we propose to incorporate the recommendation-specific DAG learner into a novel causal preference-based recommendation framework named CausPref, mainly consisting of causal learning of invariant user preference and anti-preference negative sampling to deal with implicit feedback. Extensive experimental results from real-world datasets clearly demonstrate that our approach surpasses the benchmark models significantly under types of out-of-distribution settings, and show its impressive interpretability.