Papers by Wenxuan Wang

Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language Models (2026.acl-long)

Challenge:	Large Language Models (LLMs) are transforming healthcare through their ability to understand and assist with medical tasks.
Approach:	They analyze system profiles, clinical planning, medical reasoning frameworks, and external capacity enhancement.
Outcome:	The findings highlight the future directions in medical reasoning, physical system integration, and training simulations.

Copied to clipboard

Wenxuan Wang, Zizhan Ma, Guo Yu, Yiu-Fai Cheung, Meidan Ding, Jie Liu, Wenting Chen, Linlin Shen

Hit the Nail on the Head: Parameter-Efficient Multi-task Tuning via Human Language Intervention (2024.findings-emnlp)

Challenge:	Large language models (LLMs) are proving significant potential in healthcare, prompting numerous benchmarks to evaluate their capabilities.
Approach:	They propose a framework that deconstructs benchmark development into five stages from design to governance and provides a checklist of 46 medically-tailored criteria.
Outcome:	The framework deconstructs benchmark development into five stages from design to governance and provides a comprehensive checklist of 46 medically-tailored criteria.

Copied to clipboard

Wenxuan Lu, Songhao Jiang, Wang Yijing, Tianning Zang

Decoupled Reasoning with Implicit Fact Tokens (DRIFT): A Dual-Model Framework for Efficient Long-Context Inference (2026.findings-acl)

Challenge:	Recent studies show that PEFT on small pre-trained language models improves multitasking capabilities.
Approach:	They propose a multi-task learning framework that enables transfer of prior knowledge across tasks . they attach task descriptions to input samples and map them to task embeddings .
Outcome:	The proposed method improves performance on a T5 model and in decoder-only models .

Copied to clipboard

Wenxuan Xie, Xuhong Wang

ToolSafety: A Comprehensive Dataset for Enhancing Safety in LLM-Based Agent Tool Invocations (2025.emnlp-main)

Challenge:	Existing solutions to integrate extensive, dynamic knowledge into Large Language Models (LLMs) are constrained by finite context windows, retriever noise, or the risk of catastrophic forgetting.
Approach:	They propose a dual-model architecture that explicitly decouples knowledge extraction from the reasoning process by compressing document chunks into implicit fact tokens conditioned on the query.
Outcome:	The proposed architecture significantly outperforms strong baselines among comparably sized models on long-context tasks while maintaining inference accuracy.

Copied to clipboard

Yuejin Xie, Youliang Yuan, Wenxuan Wang, Fan Mo, Jianmin Guo, Pinjia He

Challenge:	Current models exhibit notable vulnerabilities in maintaining safety during multi-step tool interactions and in indirect harm scenarios.
Approach:	They propose a safety fine-tuning dataset to fine- tune LLMs into assistants . they propose to use synthesized trajectories and realistic, context-aware sample generation .
Outcome:	The proposed model maintains safety in multi-step and indirect harm scenarios with little impact on helpfulness.

POLYCHARTQA: Benchmarking Large Vision-Language Models with Multilingual Chart Question Answering (2026.acl-long)

Copied to clipboard

Yichen Xu, Liangyu Chen, Liang Zhang, Zihao Yue, Jianzhe Ma, Wenxuan Wang, Qin Jin

Multilinguality

Challenge:	Existing chart understanding benchmarks are overwhelmingly English-centric, limiting their accessibility and relevance to global audiences.
Approach:	They propose a multilingual chart question answering benchmark that enables efficient multilingual generation via data translation and code reuse.
Outcome:	The proposed benchmark systematically evaluates multilingual chart understanding on state-of-the-art LVLMs and shows a significant performance gap between English and other languages.

Where Fact Ends and Fairness Begins: Redefining AI Bias Evaluation through Cognitive Biases (2025.findings-emnlp)

Copied to clipboard

Jen-tse Huang, Yuhang Yan, Linqi Liu, Yixin Wan, Wenxuan Wang, Kai-Wei Chang, Michael R. Lyu

ParroT: Translating during Chat using Large Language Models tuned with Human Translation and Feedback (2023.findings-emnlp)

Challenge:	Existing benchmarks conflate factual correctness and normative fairness . a model may generate responses that are factually accurate but socially unfair .
Approach:	They propose a benchmark to examine the boundary between fact and fair . they draw on representativeness bias, attribution bias and ingroup–outgroup bias to explain why models often misalign fact and faireness.
Outcome:	The proposed model is based on ten frontier models and is available on github . it is compared with a standard model that generates people of color in Nazi-era uniforms .

Copied to clipboard

Wenxiang Jiao, Jen-tse Huang, Wenxuan Wang, Zhiwei He, Tian Liang, Xing Wang, Shuming Shi, Zhaopeng Tu

Beyond Literal Descriptions: Understanding and Locating Open-World Objects Aligned with Human Intentions (2024.findings-acl)

Challenge:	Large language models (LLMs) like ChatGPT are only accessible through restricted APIs, which creates barriers to new research and advancements in the field.
Approach:	They propose a framework to enhance and regulate the translation abilities during chat . they reformulate translation data into the instruction-following style and introduce a "Hint" field .
Outcome:	The proposed framework enhances and regulates the translation abilities during chat . it reformulates translation data into the instruction-following style and introduces a "Hint" field .

Copied to clipboard

Wenxuan Wang, Yisi Zhang, Xingjian He, Yichen Yan, Zijia Zhao, Xinlong Wang, Jing Liu

Visual Data in NLP

Challenge:	Existing methods for visual grounding rely on the assumption that the given expression must be literal . this impedes the practical deployment of agents in real-world scenarios.
Approach:	They propose a visual grounding task that uses intention expressions to locate foreground entities . they build a large-scale IVG dataset with free-form intention expression to promote VG .
Outcome:	The proposed method is based on a large-scale intention-driven visual-language (V-L) dataset with free-form intention expressions.

Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training (2025.acl-long)

Copied to clipboard

Youliang Yuan, Wenxiang Jiao, Wenxuan Wang, Jen-tse Huang, Jiahao Xu, Tian Liang, Pinjia He, Zhaopeng Tu

Chain-of-Jailbreak Attack for Image Generation Models via Step by Step Editing (2025.findings-acl)

Challenge:	Large Language Models exhibit a level of intelligence that is both impressive and everevolving, but their ability to refuse generating unsafe content is a double-edged sword.
Approach:	They propose a method to tackle a refusal position bias within safety tuning data that compromises the models’ ability to appropriately refuse generating unsafe content.
Outcome:	The proposed method significantly improves model safety without compromising performance and surpasses baseline methods in defending against attacks.

Copied to clipboard

Wenxuan Wang, Kuiyi Gao, Youliang Yuan, Jen-tse Huang, Qiuzhi Liu, Shuai Wang, Wenxiang Jiao, Zhaopeng Tu

Can’t See the Forest for the Trees: Benchmarking Multimodal Safety Awareness for Multimodal LLMs (2025.acl-long)

Challenge:	Text-based image generation models, such as Stable Diffusion and DALL-E 3, hold significant potential in content creation and publishing workflows . however, considerable efforts are being made to prevent the generation of harmful content, such abusive, violent, or pornographic material.
Approach:	They propose a chain-of-jailbreak method which decomposes malicious queries into multiple sub-queries and iteratively edits images based on these sub-questions.
Outcome:	The proposed method can bypass safeguards of image generation models for over 60% cases, significantly outperforms other jailbreaking methods (14%)

Copied to clipboard

Wenxuan Wang, Xiaoyuan Liu, Kuiyi Gao, Jen-tse Huang, Youliang Yuan, Pinjia He, Shuai Wang, Zhaopeng Tu

On the Reliability of Psychological Scales on Large Language Models (2024.emnlp-main)

Challenge:	Multimodal Large Language Models (MLLMs) have expanded the capabilities of traditional language models by enabling interaction through both text and images.
Approach:	They propose a multimodal safety awareness benchmark to evaluate MLLMs across 29 safety scenarios with 1,500 carefully curated image-prompt pairs.
Outcome:	The proposed model is able to identify unsafe content and avoid over-sensitivity that can hinder helpfulness.

Copied to clipboard

Jen-tse Huang, Wenxiang Jiao, Man Ho Lam, Eric John Li, Wenxuan Wang, Michael Lyu

On the Role of Entity and Event Level Conceptualization in Generalizable Reasoning: A Survey of Tasks, Methods, Applications, and Future Directions (2025.findings-emnlp)

Challenge:	Recent research has focused on examining Large Language Models’ characteristics from a psychological standpoint, acknowledging the necessity of understanding their behavioral characteristics.
Approach:	They propose to examine the reliability of personality tests to LLMs by using psychological scales.
Outcome:	The proposed model can represent diverse personalities with specific prompt instructions.

Copied to clipboard

Weiqi Wang, Tianqing Fang, Haochen Shi, Baixuan Xu, Wenxuan Ding, Liyu Zhang, Wei Fan, Jiaxin Bai, Haoran Li, Xin Liu, Yangqiu Song

MlingConf: A Comprehensive Study of Multilingual Confidence Estimation on Large Language Models (2025.findings-acl)

Challenge:	Conceptualization is a fundamental element of human cognition and plays a pivotal role in generalizable reasoning.
Approach:	They propose to categorize different types of conceptualizations into four levels based on the types of instances being conceptualized.
Outcome:	The proposed categorization of different types of conceptualizations into four levels based on the types of instances being conceptualized .

Copied to clipboard

Boyang Xue, Hongru Wang, Rui Wang, Sheng Wang, Zezhong Wang, Yiming Du, Bin Liang, Wenxuan Zhang, Kam-Fai Wong

Learning Auxiliary Tasks Improves Reference-Free Hallucination Detection in Open-Domain Long-Form Generation (2025.acl-short)

Challenge:	Existing studies on LLM confidence estimations in languages other than English have been limited to English.
Approach:	They propose to use question-related language to prompt LLMs to assess their confidence in large language models.
Outcome:	The proposed model improves on question-related language prompts for LS tasks, while English exhibits notable linguistic dominance in confidence estimations.

Copied to clipboard

Chengwei Qin, Wenxuan Zhou, Karthik Abinav Sankararaman, Nanshu Wang, Tengyu Xu, Alexander Radovic, Eryk Helenowski, Arya Talebzadeh, Aditya Tayade, Sinong Wang, Shafiq Joty, Han Fang, Hao Ma

Exploring Attention Attractors in Large Language Models (2026.acl-long)

Challenge:	Existing methods for detecting hallucination in long-form tasks focus on limited domains or rely heavily on external fact-checking tools, which may not always be available.
Approach:	They propose a new paradigm that augments fine-tuning with an auxiliary task for the model to jointly learn with the main task of hallucination detection.
Outcome:	The proposed method outperforms existing methods for detecting hallucination in open-domain long-form generation and is more accurate than random guessing.

Copied to clipboard

Ziheng Wang, Zihao Yue, Wenxuan Wang, Qin Jin

Reframe Your Life Story: Interactive Narrative Therapist and Innovative Moment Assessment with Large Language Models (2025.emnlp-main)

Challenge:	Existing studies have suggested that attention attractors function as "summary tokens" while others speculate that tokens with weaker semantics attract high attention, they act as attention sinks that offload excessive attention.
Approach:	They examine attention attractors, tokens that draw significantly high attention, in large language models.
Outcome:	The proposed models are able to capture long-range dependencies within a given context.

Copied to clipboard

Yi Feng, Jiaqi Wang, Wenxuan Zhang, Zhuang Chen, Shen Yutong, Xiyao Xiao, Minlie Huang, Liping Jing, Jian Yu

A Causal View of Entity Bias in (Large) Language Models (2023.findings-emnlp)

Challenge:	Existing approaches to mental health support lack realism and capture therapeutic progression over time.
Approach:	They propose a framework that simulates expert narrative therapists by planning therapeutic stages, guiding reflection levels, and generating contextually appropriate responses through retrieval-augmentation.
Outcome:	The proposed framework outperforms standard methods in quality and depth on 260 simulated clients and 230 human participants.

Copied to clipboard

Fei Wang, Wenjie Mo, Yiwei Wang, Wenxuan Zhou, Muhao Chen

Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation (2026.findings-acl)

Challenge:	Entity bias affects pretrained (large) language models, causing them to rely on (biased) parametric knowledge to make unfaithful predictions.
Approach:	They propose a structured causal model whose parameters are easier to estimate . they propose to perturb the original entity with neighboring entities .
Outcome:	The proposed model reduces biasing information pertaining to the original entity while still preserving sufficient semantic information from similar entities.

Copied to clipboard

Jen-tse Huang, Chang Chen, Shiyang Lai, Wenxuan Wang, Michelle R Kaufman, Mark Dredze

AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor (2026.findings-acl)

Challenge:	Short-video platforms have become major channels for misinformation, but their robustness against misinformation entangled with cognitive biases remains under-explored.
Approach:	They propose a framework for evaluation of short-video platforms that use visual cues and social cue.
Outcome:	The proposed framework evaluates MLLMs across five modality settings.

Copied to clipboard

Shu Yang, Jingyu Hu, Tong Li, Hanqi Yan, Wenxuan Wang, Di Wang

Challenge:	AutoMonitor-Bench evaluates the reliability of LLM-based misbehavior monitors across diverse tasks and failure modes.
Approach:	They introduce AutoMonitor-Bench, a benchmark designed to evaluate misbehavior monitors across diverse tasks and failure modes.
Outcome:	The new benchmark evaluates the reliability of LLM-based misbehavior monitors across tasks and failure modes.

MetaScale: Test-Time Scaling with Evolving Meta-Thoughts (2026.findings-acl)

Copied to clipboard

Qin Liu, Wenxuan Zhou, Nan Xu, James Y. Huang, Fei Wang, Sheng Zhang, Hoifung Poon, Muhao Chen

QADYNAMICS: Training Dynamics-Driven Synthetic QA Diagnostic for Zero-Shot Commonsense Question Answering (2023.findings-emnlp)

Challenge:	Existing approaches impose fixed cognitive structures that enhance performance in specific tasks but lack adaptability across diverse scenarios.
Approach:	They propose a test-time scaling framework based on meta-thoughts to improve performance . meta-thinkts are adaptive thinking strategies tailored to a given task .
Outcome:	Experimental results show that MetaScale outperforms standard inference approaches . it can scale more effectively with increasing sampling budgets and produces more structured responses .

Copied to clipboard

Haochen Shi, Weiqi Wang, Tianqing Fang, Baixuan Xu, Wenxuan Ding, Xin Liu, Yangqiu Song

LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models (2024.emnlp-main)

Challenge:	Existing approaches to QA fine-tune language models on QA pairs constructed from CommonSense Knowledge Bases (CSKBs) however, current QA synthesis protocols introduce noise from the CSKB and generate ungrammatical questions and false negative options, which impede the model’s ability to generalize.
Approach:	They propose a framework to analyze the training dynamics of each QA pair at both the question level and option level, discarding machine-detectable artifacts and mislabeled or false-negative options.
Outcome:	The proposed framework outperforms baseline approaches while using only 33% of the synthetic data.

Copied to clipboard

Yuxuan Wan, Wenxuan Wang, Yiliu Yang, Youliang Yuan, Jen-tse Huang, Pinjia He, Wenxiang Jiao, Michael Lyu

Global Eye: Breaking the “Fixed Thinking Pattern” during the Instruction Expansion Process (2025.acl-long)

Challenge:	LogicAsker examines and improves the reasoning abilities of large language models such as ChatGPT and GPT-4.
Approach:	They propose a set of atomic reasoning skills grounded in propositional and predicate logic to examine and improve the reasoning abilities of large language models such as ChatGPT and GPT-4.
Outcome:	The proposed approach improves reasoning abilities in large language models such as ChatGPT and GPT-4 by up to 5%.

Copied to clipboard

Wenxuan Lu, Wei Liu, Jian Luan, Bin Wang, Songhao Jiang, Tianning Zang

Should We Rely on Entity Mentions for Relation Extraction? Debiasing Relation Extraction with Counterfactual Analysis (2022.naacl-main)

Challenge:	Existing methods focus on constructing multi-perspective prompts to expand instructions, overlooking the “Fixed Thinking Pattern” issue of Large Language Models.
Approach:	They propose a method that analyzes the statistical characteristics of newly generated instructions and updates the prompts after a fixed number of instruction expansions.
Outcome:	The proposed method surpasses open-source LLMs and GPT3.5 in several metrics.

Copied to clipboard

Yiwei Wang, Muhao Chen, Wenxuan Zhou, Yujun Cai, Yuxuan Liang, Dayiheng Liu, Baosong Yang, Juncheng Liu, Bryan Hooi

Information Extraction & Text Mining

Challenge:	Existing studies rely on entity information for sentence-level relation extraction (RE) but this can leak superficial and spurious clues of relations.
Approach:	They propose to use entity mentions to extract relations from textual context . they use a causal graph to model dependencies between variables in RE models .
Outcome:	The proposed method yields significant gains on both effectiveness and generalization for RE.

JARVIS or Ultron? A Survey on the Safety and Security Threats of Computer-Using Agents (2026.acl-long)

Copied to clipboard

Ada Chen, Yongjiang Wu, Junyuan Zhang, Jingyu Xiao, Shu Yang, Jen-tse Huang, Kun Wang, Wenxuan Wang, Shuai Wang

Challenge:	Recent advances in computer-using agents have created new safety and security risks . despite the impressive capabilities of CUAs, there are still significant security risks.
Approach:	They propose a systematization of knowledge on the safety and security threats of Computer-Using Agents.
Outcome:	The proposed framework provides a framework for assessing the safety and security risks of computer-using agents.

Social Welfare Function Leaderboard: On the Emergence of LLM Agents as the Welfare Dictator (2026.findings-acl)

Copied to clipboard

Zhengliang Shi, Ruotian Ma, Jen-tse Huang, Xinbei Ma, Xingyu Chen, Mengru Wang, Qu Yang, Yue Wang, Fanghua Ye, Ziyang Chen, Shanyi Wang, Cixing LI, Wenxuan Wang, Zhaopeng Tu, Xiaolong Li, Zhaochun Ren, Liefeng Bo

AdaMergeX: Cross-Lingual Transfer with Large Language Models via Adaptive Adapter Merging (2025.naacl-long)

Challenge:	Large language models (LLMs) are increasingly entrusted with high-stakes decisions that affect human welfare.
Approach:	They evaluate 20 state-of-the-art Large language models (LLMs) and 20 LLM dictators to create a social welfare function benchmark.
Outcome:	The proposed model creates dilemma between maximizing collective efficiency and ensuring distributive fairness.

Copied to clipboard

Yiran Zhao, Wenxuan Zhang, Huiming Wang, Kenji Kawaguchi, Lidong Bing

KnowCoder-X: Boosting Multilingual Information Extraction via Code (2025.findings-acl)

Challenge:	Large Language Models excel in highresource languages but underperform in lowresource ones.
Approach:	They propose a cross-lingual transfer method that decouples "task ability" from "language ability" they propose to use adaptive adapter merging to obtain target adapters by combining other adapters.
Outcome:	The proposed method outperforms existing methods in highresource languages . it decouples "task ability" from "language ability" but fails to fully separate "task capability" from the "source language"

Copied to clipboard

Yuxin Zuo, Wenxuan Jiang, Wenxuan Liu, Zixuan Li, Long Bai, Hanbin Wang, Yutao Zeng, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng

Multilinguality

Challenge:	Empirical evidence indicates that Large Language Models exhibit spontaneous cross-lingual alignment in Information Extraction (IE) however, a significant imbalance across languages persists, highlighting an underlying deficiency.
Approach:	They propose a code LLM with advanced cross-lingual and multilingual capabilities for universal IE that standardizes the representation of multilingual schemas using Python classes and conducts IE alignment instruction tuning on translated instance prediction task.
Outcome:	The proposed model surpasses ChatGPT and SoTA by 30.17% without training in 29 unseen languages and significantly improves cross-lingual IE transferability.

Improving Knowledge Graph Completion with Structure-Aware Supervised Contrastive Learning (2024.emnlp-main)

Copied to clipboard

Jiashi Lin, Lifang Wang, Xinyu Lu, Zhongtian Hu, Wei Zhang, Wenxuan Lu

Identifying the Achilles’ Heel: An Iterative Method for Uncovering Factual Errors in Large Language Models (2026.findings-acl)

Challenge:	Existing contrastive methods focus on individual triples, overlooking the broader structural connectivities and topologies of KGs.
Approach:	They propose a new contrastive learning framework that incorporates four tasks specifically tailored to KG data: Vertex-level CL, Neighbor-level Cl, Path-levelCL, and Relation composition level CL.
Outcome:	The proposed framework achieves SOTA performance under standard supervised and low-resource settings.

Copied to clipboard

Wenxuan Wang, Yuk-Kit Chan, Zixuan Ling, Shi Juluan, Youliang Yuan, Jen-tse Huang, Yifei Zhang, Wenxiang Jiao, Zhaopeng Tu, Michael R. Lyu

Order-Agnostic Data Augmentation for Few-Shot Named Entity Recognition (2024.acl-long)

Challenge:	Current methods for evaluating LLMs’ veracity are limited by the need for extensive human labor, test data contamination, or limited scope, hindering efficient and effective exposure of errors.
Approach:	They propose a framework that extracts fact triplets to generate diverse question types using rule-based natural language processing techniques.
Outcome:	The proposed framework can trigger factual errors in up to 55% of questions in large LLMs while maintaining coverage of questions.

Copied to clipboard

Huiming Wang, Liying Cheng, Wenxuan Zhang, De Wen Soh, Lidong Bing

Low-Resource NLP

Challenge:	Existing DA methods for named entity recognition (NER) are costly and labor-intensive to acquire, necessitating innovative approaches to data scarcity.
Approach:	They propose an order-agnostic data augmentation solution that exploits the order-based property in the training phase of sequence-to-sequence NER methods for data augmented.
Outcome:	The proposed method significantly enhances the few-shot capabilities of pre-trained language models in low-resource settings.

LongMP-Bench: A Benchmark for Multimodal Persona Understanding in Long-Term Dialogues (2026.findings-acl)

Copied to clipboard

Zhuoqun Li, Zhaopei Huang, Wenxuan Wang, Qin Jin

Does ChatGPT Know That It Does Not Know? Evaluating the Black-Box Calibration of ChatGPT (2024.lrec-main)

Challenge:	Existing datasets suffer from limited persona diversity and static, overly simplified settings, making them insufficient for capturing the complexity of real-world interactions.
Approach:	They propose a benchmark to evaluate models' ability to understand evolving user personas within long-term multimodal dialogues by using a dataset that contains long conversations from 150 users.
Outcome:	The proposed benchmark aims to assess models' ability to track persona evolution, integrate visual and textual inputs, and apply persona understanding in realistic dialogue scenarios.

Copied to clipboard

Youliang Yuan, Wenxuan Wang, Qingshuo Guo, Yiming Xiong, Chihao Shen, Pinjia He

Natural Language Interfaces

Challenge:	Recent performance of ChatGPT in downstream tasks is questionable, but does it know that it does not know?
Approach:	They propose to use three types of proxy confidence to evaluate ChatGPT's black-box calibration ability.
Outcome:	The proposed model exhibits a positive correlation with accuracy in TruthfulQA and a negative correlation in the ModAr dataset.

Learning to Ask: When LLM Agents Meet Unclear Instruction (2025.emnlp-main)

Copied to clipboard

Wenxuan Wang, Shi Juluan, Zixuan Ling, Yuk-Kit Chan, Chaozheng Wang, Cheryl Lee, Youliang Yuan, Jen-tse Huang, Wenxiang Jiao, Michael R. Lyu

Challenge:	Despite their impressive capabilities, LLMs struggle with complex computations and delivering accurate, timely information.
Approach:	They propose a framework that prompts LLM agents to ask questions when they encounter obstacles due to unclear instructions and an automated evaluation tool called ToolEvaluator.
Outcome:	The proposed framework outperforms existing frameworks for tool learning in the Noisy ToolBench.

mDPO: Conditional Preference Optimization for Multimodal Large Language Models (2024.emnlp-main)

Copied to clipboard

Fei Wang, Wenxuan Zhou, James Y. Huang, Nan Xu, Sheng Zhang, Hoifung Poon, Muhao Chen

Improving Model Factuality with Fine-grained Critique-based Evaluator (2025.acl-long)

Challenge:	Recent studies have attempted to apply DPO to multimodal scenarios but have found it challenging to achieve consistent improvement.
Approach:	They propose a multimodal DPO objective that prevents the over-prioritization of language-only preferences by also optimizing image preference.
Outcome:	The proposed method significantly improves performance on two multimodal LLMs of different sizes and three widely used benchmarks.

Copied to clipboard

Yiqing Xie, Wenxuan Zhou, Pradyot Prakash, Di Jin, Yuning Mao, Quintin Fettes, Arya Talebzadeh, Sinong Wang, Han Fang, Carolyn Rose, Daniel Fried, Hejia Zhang

AI Sees Your Location—But With A Bias Toward The Wealthy World (2025.emnlp-main)

Challenge:	Factuality evaluation aims to detect factual errors produced by language models and guide the development of more factual models.
Approach:	They propose a framework that leverages FenCE to improve the factuality of LM generators by constructing training data.
Outcome:	The proposed framework improves the factuality of LM generators by enhancing their training data.

Copied to clipboard

Jingyuan Huang, Jen-tse Huang, Ziyi Liu, Xiaoyuan Liu, Wenxuan Wang, Jieyu Zhao

Rethinking the Value of Transformer Components (2020.coling-main)

Challenge:	Visual Language Models (VLMs) have shown remarkable performance across various tasks, particularly in recognizing geographic information from images.
Approach:	They propose to use 1,200 images paired with detailed geographic metadata to evaluate VLMs' performance.
Outcome:	The models achieve 53.8% accuracy in city prediction, but exhibit significant biases in regional tasks.

Copied to clipboard

Wenxuan Wang, Zhaopeng Tu

Curing Miracle Steps in LLM Mathematical Reasoning with Rubric Rewards (2026.acl-long)

Challenge:	Empirical results show that certain components are more important than others . we propose a new training strategy that can improve Transformer models by distinguishing unimportant components .
Approach:	They propose a training strategy that distinguishes the unimportant components in training . they compare the impact of individual component (sub-layer) on model performance .
Outcome:	The proposed training strategy can improve translation performance by distinguishing unimportant components in training.

Copied to clipboard

Youliang Yuan, Qiuyang Mang, Jingbang Chen, Hong Wan, Xiaoyuan Liu, Junjielong Xu, Jen-tse Huang, Wenxuan Wang, Wenxiang Jiao, Pinjia He

Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models (2024.acl-long)

Challenge:	Existing models are susceptible to reward hacking, leading to a substantial overestimation of a model's reasoning ability.
Approach:	They propose a Rubric Reward Model that rewards the entire reasoning trajectory against problem-specific rubrics.
Outcome:	The proposed model outperforms outcome-only supervision on four math benchmarks and boosts Verified Pass@1024 from 26.7% to 62.6% and reduces the incidence of Miracle Steps by 71%.

Copied to clipboard

Wenxuan Wang, Wenxiang Jiao, Jingyuan Huang, Ruyi Dai, Jen-tse Huang, Zhaopeng Tu, Michael Lyu

MedEinst: Benchmarking the Einstellung Effect in Medical LLMs through Counterfactual Differential Diagnosis (2026.acl-long)

Challenge:	e.g., ChatGPT often provides inappropriate English-culture-related answers when users ask in non-English languages.
Approach:	They build a benchmark of concrete and abstract cultural objects to evaluate the cultural dominance issue in large language models.
Outcome:	The proposed model can significantly mitigate cultural dominance issue in large language models . the model can provide accurate answers in English, while the model is ethically sound .

Copied to clipboard

Wenting Chen, Guolin Huang, Wenxuan Wang, Zhongrui Zhu

Structured Data in NLP

Challenge:	Existing medical benchmarks fail to detect the Einstellung Effect in clinical diagnosis . Existing models exhibit the Einstellung effect, relying on statistical shortcuts rather than logical reasoning.
Approach:	They propose a counterfactual benchmark that uses statistical shortcuts to diagnose patients . they propose CGME-based system that iteratively refines reasoning paths .
Outcome:	The proposed model achieves high baseline accuracy but severe bias trap rates . iteratively refines reasoning paths in an exemplar base and consolidates disease-specific knowledge into illness graphs.

A Survey of Deep Learning for Geometry Problem Solving (2026.acl-long)

Copied to clipboard

Jianzhe Ma, Wenxuan Wang, Qin Jin

Contrastive Instruction Tuning (2024.findings-acl)

Challenge:	Recent surge in deep learning technologies has significantly accelerated research in this area.
Approach:	They propose a comprehensive summary of the relevant tasks in geometry problem solving and a review of related deep learning methods.
Outcome:	The proposed method is based on a systematic review of related methods and evaluation metrics and methods.

Copied to clipboard

Tianyi Yan, Fei Wang, James Y. Huang, Wenxuan Zhou, Fan Yin, Aram Galstyan, Wenpeng Yin, Muhao Chen

Entity-centered Cross-document Relation Extraction (2022.emnlp-main)

Challenge:	Current LLMs exhibit limited robustness to unseen instructions, generating inconsistent outputs when the same instruction is phrased with slightly varied forms or language styles.
Approach:	They propose a method which maximizes the similarity between the hidden representations of semantically equivalent instruction-instance pairs while minimizing the similarities between semantically different ones.
Outcome:	Experiments on the PromptBench benchmark show that Contrastive Instruction Tuning improves LLMs’ robustness to unseen instructions with variations across character, word, sentence, and semantic levels by +2.5% in accuracy.

Copied to clipboard

Fengqi Wang, Fei Li, Hao Fei, Jingye Li, Shengqiong Wu, Fangfang Su, Wenxuan Shi, Donghong Ji, Bo Cai

Information Extraction & Text Mining

Challenge:	Existing methods for relation extraction only use text snippets surrounding target entities in multiple documents.
Approach:	They propose a relation-extraction model that uses cross-path entity relation attention to detect the semantic relations between entities in a given text.
Outcome:	The proposed method outperforms the state-of-the-art methods in the dataset CodRED by 10%.

VisCRA: A Visual Chain Reasoning Attack for Jailbreaking Multimodal Large Language Models (2025.emnlp-main)

Copied to clipboard

Bingrui Sima, Linhua Cong, Wenxuan Wang, Kun He

Robustness in NLP

Challenge:	Recent advances in Large Reasoning Models (LRMs) have enabled sophisticated visual reasoning capabilities by integrating reinforcement learning and Chain-of-Thought (CoT) supervision.
Approach:	They propose a jailbreak framework that exploits visual reasoning chains to bypass safety mechanisms.
Outcome:	The proposed framework achieves high attack success rates on leading closed-source MLRMs.

UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated Supervision (2026.acl-long)

Copied to clipboard

Zhen Fang, Ruiyan Han, XinYu Sun, Yuchen Ma, Ziheng Wang, Yu Zeng, Zehui Chen, Lin Chen, Wenxuan Huang, Wei-Jie Xu, Yi Cao, Feng Zhao

Getting Sick After Seeing a Doctor? Diagnosing and Mitigating Knowledge Conflicts in Event Temporal Reasoning (2024.findings-naacl)

Challenge:	Unified Multimodal Models have achieved remarkable success in cross-modal comprehension, but a gap persists in their ability to translate internal knowledge into faithful and controllable synthesis.
Approach:	They propose a self-improvement framework that partitions a single UMM into three collaborative roles: Proposer, Solver, and Judge.
Outcome:	The proposed framework improves on TIIF, DPG, CompBench and UniCycle benchmarks.

Copied to clipboard

Tianqing Fang, Zhaowei Wang, Wenxuan Zhou, Hongming Zhang, Yangqiu Song, Muhao Chen

SeaLLMs - Large Language Models for Southeast Asia (2024.acl-demos)

Challenge:	Event temporal reasoning aims at identifying the temporal relations between two or more events from narratives.
Approach:	They propose to detect knowledge conflicts in event temporal reasoning using bias indicators such as event relation prior bias, tense bias, narrative bias, and dependency bias.
Outcome:	The proposed method can be applied to Pre-trained Language Models and Large Language Model (LLMs) as additional training data or demonstrations for In- Context Learning.

Copied to clipboard

Xuan-Phi Nguyen, Wenxuan Zhang, Xin Li, Mahani Aljunied, Zhiqiang Hu, Chenhui Shen, Yew Ken Chia, Xingxuan Li, Jianyu Wang, Qingyu Tan, Liying Cheng, Guanzheng Chen, Yue Deng, Sen Yang, Chaoqun Liu, Hang Zhang, Lidong Bing

When One LLM Drools, Multi-LLM Collaboration Rules (2026.acl-long)

Challenge:	Existing large language models favor high-resource languages, such as English, at the expense of low-resourced and regional languages.
Approach:	They propose a series of language models that specifically focuses on Southeast Asian languages.
Outcome:	SeaLLM models outperform ChatGPT-3.5 in non-Latin languages by large margins . linguistic disparity impedes access to state-of-the-art AI technologies for non-English-speaking populations .

Copied to clipboard

Shangbin Feng, Wenxuan Ding, Alisa Liu, Zifeng Wang, Weijia Shi, Yike Wang, Shannon Zejiang Shen, Xiaochuang Han, Hunter Lang, Chen-Yu Lee, Tomas Pfister, Yejin Choi, Yulia Tsvetkov

Challenge:	a single general-purpose LLM is not enough to produce a reliable output, argues this paper . a multi-LLM collaboration approach addresses reliability, democratization, and pluralism .
Approach:	They argue that a single general-purpose LLM is not enough to produce a reliable output . they organize existing multi-LLM collaboration methods into a hierarchy based on access and information exchange .
Outcome:	The proposed method addresses reliability, democratization, and pluralism challenges a single LLM fails to produce a reliable output.

MIND: Multimodal Shopping Intention Distillation from Large Vision-language Models for E-commerce Purchase Understanding (2024.emnlp-main)

Copied to clipboard

Baixuan Xu, Weiqi Wang, Haochen Shi, Wenxuan Ding, Huihao Jing, Tianqing Fang, Jiaxin Bai, Xin Liu, Changlong Yu, Zheng Li, Chen Luo, Qingyu Yin, Bing Yin, Long Chen, Yangqiu Song

Green & Sustainable NLP

Challenge:	Existing methods for acquiring large-scale intentions generate product-centric intentions without product images and incur high costs for scalability.
Approach:	They propose a multimodal framework that allows Large Vision-Language Models to infer purchase intentions from multimodal product metadata and prioritize human-centric ones.
Outcome:	The proposed framework shows that it is robust to different prompts and superior to previous methods.

Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification (2026.findings-acl)

Copied to clipboard

Yuxuan Wan, Tianqing Fang, Zaitang LI, Yintong Huo, Wenxuan Wang, Haitao Mi, Dong Yu, Michael R. Lyu

Challenge:	Recent advances in Deep Research Agents (DRAs) are transforming automated knowledge discovery and problem-solving.
Approach:	They propose an inference-time scaling of verification wherein an agent self-improves at test time by evaluating its generated answers.
Outcome:	The proposed model outperforms vanilla agent-as-judge and LLM judge baselines by 12%–48% in meta-evaluation F1 score.

Robust Natural Language Understanding with Residual Attention Debiasing (2023.findings-acl)

Copied to clipboard

Fei Wang, James Y. Huang, Tianyi Yan, Wenxuan Zhou, Muhao Chen

Robustness in NLP

Challenge:	Existing ensemble-based debiasing methods do not address unintended dataset biases . attention plays a crucial role in providing robust prediction in NLU models .
Approach:	They propose an end-to-end debiasing method that mitigates unintended biases from attention.
Outcome:	The proposed method improves the OOD performance of BERT-based models on three benchmarks.

VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models (2025.emnlp-main)

Copied to clipboard

Jen-tse Huang, Jiantong Qin, Jianping Zhang, Youliang Yuan, Wenxuan Wang, Jieyu Zhao

Visual Data in NLP

Challenge:	Identifying and addressing potential social biases is essential to prevent harm to users.
Approach:	They examine explicit and implicit biases exhibited by Vision-Language Models . they pose questions related to gender and racial differences to test their models .
Outcome:	The proposed models are used in image description tasks, form completion tasks and medical applications.

Teaching LLMs to Abstain across Languages via Multilingual Feedback (2024.emnlp-main)

Copied to clipboard

Shangbin Feng, Weijia Shi, Yike Wang, Wenxuan Ding, Orevaoghene Ahia, Shuyue Stella Li, Vidhisha Balachandran, Sunayana Sitaram, Yulia Tsvetkov

Multilinguality

Challenge:	Existing studies on LLM abstention focus on English, but they show that it can reduce the accuracy of the model by 20.5% .
Approach:	They propose to teach LLMs to abstain in the face of knowledge gaps by generating multiple feedback items in related languages.
Outcome:	Extensive experiments show that the proposed approach outperforms baselines and achieves 9.2% improvement for low-resource languages.

IntentionQA: A Benchmark for Evaluating Purchase Intention Comprehension Abilities of Language Models in E-commerce (2024.findings-emnlp)

Copied to clipboard

Wenxuan Ding, Weiqi Wang, Sze Kwok, Minghao Liu, Tianqing Fang, Jiaxin Bai, Xin Liu, Changlong Yu, Zheng Li, Chen Luo, Qingyu Yin, Bing Yin, Junxian He, Yangqiu Song

EngiBench: A Benchmark for Evaluating Large Language Models on Engineering Problem Solving (2026.findings-acl)

Challenge:	Existing approaches that distill intentions from LMs fail to generate meaningful and human-centric intentions applicable in real-world E-commerce contexts.
Approach:	They propose a double-task multiple-choice question answering benchmark to evaluate LMs' comprehension of purchase intentions in E-commerce.
Outcome:	The proposed benchmark consists of 4,360 carefully curated problems across three difficulty levels, constructed using an automated pipeline to ensure scalability on large E-commerce platforms.

Copied to clipboard

Xiyuan Zhou, Xinlei Wang, Yirui He, Ruixi Zou, Yang Wu, Yuheng Cheng, Yulu Xie, Wenxuan Liu, Huan Zhao, Yan Xu, Jinjin Gu, Junhua Zhao

Understanding and Improving Sequence-to-Sequence Pretraining for Neural Machine Translation (2022.acl-long)

Challenge:	Existing benchmarks focus on well-defined or abstract reasoning and fail to capture real-world engineering problems.
Approach:	They propose a hierarchical benchmark to evaluate large language models on engineering problems.
Outcome:	The proposed model performs well under well-defined conditions and is based on three levels of difficulty and covers diverse engineering subfields.

Copied to clipboard

Wenxuan Wang, Wenxiang Jiao, Yongchang Hao, Xing Wang, Shuming Shi, Zhaopeng Tu, Michael Lyu

CAR: Conceptualization-Augmented Reasoner for Zero-Shot Commonsense Question Answering (2023.findings-emnlp)

Challenge:	Existing studies on self-supervised pretraining for machine translation have focused on the jointly pretrained decoder .
Approach:	They propose a method to improve neural machine translation by jointly pretrained decoder . they propose two strategies to remedy the domain and objective discrepancies .
Outcome:	The proposed approach improves translation performance and model robustness on three language pairs.

Copied to clipboard

Weiqi Wang, Tianqing Fang, Wenxuan Ding, Baixuan Xu, Xin Liu, Yangqiu Song, Antoine Bosselut

SlideCoder: Layout-aware RAG-enhanced Hierarchical Slide Generation from Design (2025.emnlp-main)

Challenge:	Existing approaches to zero-shot commonsense question answering use incomplete CSKBs . lack of human annotations makes sampled negative examples potentially uninformative and contradictory.
Approach:	They propose a framework that abstracts a commonsense knowledge triple to many higher-level instances, which increases the coverage of the CSKB and expands the ground-truth answer space.
Outcome:	Experiments show that CAR can generalize to zero-shot commonsense scenarios . lack of human annotations makes sampled negative examples potentially uninformative and contradictory.

Copied to clipboard

Wenxin Tang, Jingyu Xiao, Wenxuan Jiang, Xi Xiao, Yuhang Wang, Xuxin Tang, Qing Li, Yuehe Ma, Junliang Liu, Shisong Tang, Michael R. Lyu

CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning (2024.acl-long)

Challenge:	Existing natural language-based LLM generation methods struggle to capture visual and structural nuances of slide designs.
Approach:	They propose a layout-aware framework for generating editable slides from reference images . they propose python code that translates NL instructions into Python code to construct each slide .
Outcome:	The proposed framework outperforms state-of-the-art models by up to 40.5 points . it also outperformed open-source models with improved reverse-engineered data.

Copied to clipboard

Weiqi Wang, Tianqing Fang, Chunyang Li, Haochen Shi, Wenxuan Ding, Baixuan Xu, Zhaowei Wang, Jiaxin Bai, Xin Liu, Cheng Jiayang, Chunkit Chan, Yangqiu Song

False Sense of Security: Why Probing-based Malicious Input Detection Fails to Generalize (2026.findings-acl)

Challenge:	Existing approaches to generalize commonsense reasoning lack instantiated knowledge and require pre-built concept taxonomies and annotations.
Approach:	They propose a framework that iteratively performs contextualized conceptualization and instantiation over commonsense knowledge bases by instructing large language models to generate both types of knowledge with critic filtering.
Outcome:	Empirical results show that distilling CANDLE on student models provides benefits across three downstream tasks.

Copied to clipboard

Cheng Wang, Zeming Wei, Qin Liu, Wenxuan Zhou, Muhao Chen

A Survey of Large Models in Sports (2026.findings-acl)

Challenge:	Recent work has leveraged probing-based approaches to study the separability of malicious and benign inputs in Large Language Models’ internal representations.
Approach:	They propose to use probing-based methods to study separability of malicious and benign inputs in LLMs' internal representations to detect harmful and benign content.
Outcome:	The proposed methods show that they learn superficial patterns rather than semantic harmfulness.

Copied to clipboard

Yichen Xu, Jianzhe Ma, Chuhan Wang, Zhonghao Cao, Liangyu Chen, Wenxuan Wang, Qin Jin

Asclepius: A Spectrum Evaluation Benchmark for Medical Multi-Modal Large Language Models (2025.acl-long)

Challenge:	Increasing interest in sports has led to the rapid advancement of large models, particularly multimodal large language models (MLLMs) . linguistic intelligence is a key component of large-model-driven sports intelligence .
Approach:	They propose to establish a foundation for advancing research and practical development of large-model-driven sports intelligence.
Outcome:	The proposed model-driven sports intelligence will be able to process and generate sports-related language effectively and process multiple data modalities.

Copied to clipboard

Jie Liu, Wenxuan Wang, Su Yihang, Jingyuan Huang, Yudi Zhang, Cheng-Yi Li, Wenting Chen, Xiaohan Xing, Kao-Jung Chang, Linlin Shen, Michael R. Lyu

Taming Extreme Tokens: Covariance-Aware GRPO with Gaussian-Kernel Advantage Reweighting (2026.acl-short)

Challenge:	Medical Multi-Modal Large Language Models (Med-MLLMs) are a promising new form of artificial general intelligence due to their ability to tackle complex tasks.
Approach:	They propose a new benchmark that comprehensively assesses medical multi-modal large language models in terms of distinct medical specialties and different diagnostic capacities.
Outcome:	The proposed model covers 15 medical specialties and different diagnostic capacities, and excludes overlap with existing VQA dataset.

Copied to clipboard

Cheng Wang, Qin Liu, Wenxuan Zhou, Muhao Chen

IntentionESC: An Intention-Centered Framework for Enhancing Emotional Support in Dialogue Systems (2025.findings-acl)

Challenge:	Excessive exploitation can cause the model to become overconfident in its suboptimal solutions, thereby limiting its capabilities to explore novel reasoning strategies.
Approach:	They propose a method that dynamically down-weights extreme token-level updates via a Gaussian kernel and reduces the instability caused by the trade-off.
Outcome:	The proposed method improves downstream performance across reasoning benchmarks and stabilizes entropy as training progresses.

Copied to clipboard

Xinjie Zhang, Wenxuan Wang, Qin Jin

Dialogue Systems & Conversational Agents

Challenge:	IntentionESC defines the possible intentions of supporters in emotional support conversations, identifies key emotional state aspects for inferring these intentions, and maps them to appropriate support strategies.
Approach:	They propose an Intention-centered Emotional Support Conversation framework which defines the possible intentions of supporters in emotional support conversations, identifies key emotional state aspects for inferring intentions, and maps them to appropriate support strategies.
Outcome:	The proposed framework defines the possible intentions of supporters in emotional support conversations, identifies key emotional state aspects for inferring these intentions, and maps them to appropriate support strategies.

EAGLE: Expert-Guided Self-Enhancement for Preference Alignment in Pathology Large Vision-Language Model (2025.acl-long)

Copied to clipboard

Meidan Ding, Jipeng Zhang, Wenxuan Wang, Haiqin Zhong, Xiaoqin Wang, Xinheng Lyu, Wenting Chen, Linlin Shen

QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language (2025.findings-acl)

Challenge:	Recent advances in Large Vision Language Models (LVLMs) show promise for pathological diagnosis, yet their application in clinical settings faces critical challenges of multimodal hallucination and biased responses.
Approach:	They propose a framework that integrates medical expertise into preference alignment.
Outcome:	The proposed framework outperforms existing pathological LVLMs while maintaining pathological accuracy.

Copied to clipboard

Qingsong Zou, Jingyu Xiao, Qing Li, Zhi Yan, Yuhang Wang, Li Xu, Wenxuan Wang, Kuofeng Gao, Ruoyu Li, Yong Jiang

All Languages Matter: On the Multilingual Safety of LLMs (2024.findings-acl)

Challenge:	Existing methods to bypass security defenses of large language models (LLMs) are not effective, but QueryAttack can be jailbroken.
Approach:	They propose a framework to examine generalizability of safety alignment by translating malicious queries into structured non-natural query languages.
Outcome:	The proposed framework can achieve high attack success rates and jailbreak various defense methods on mainstream LLMs.

Copied to clipboard

Wenxuan Wang, Zhaopeng Tu, Chang Chen, Youliang Yuan, Jen-tse Huang, Wenxiang Jiao, Michael Lyu