Papers by Wenxuan Zhang

Hierarchical Memory Organization for Wikipedia Generation (2025.acl-long)

Challenge:	Sentiment analysis models often fail to capture the broader complexities of sentiment analysis.
Approach:	They propose a task to evaluate sentiment understanding through two subtasks . they annotate a new dataset comprising 15,028 statements from 3,638 reviews .
Outcome:	The proposed task evaluates sentiment understanding through two subtasks . it is a challenging task for both small and large language models, with performance gaps of up to 27% .

Copied to clipboard

Eugene J. Yu, Dawei Zhu, Yifan Song, Xiangyu Wong, Jiebin Zhang, Wenxuan Shi, Xiaoguang Li, Qun Liu, Sujian Li

Challenge:	Existing methods for generating Wikipedia articles do not utilize memory directly for outline generation.
Approach:	They propose a method to generate Wikipedia articles autonomously by leveraging a hierarchical memory architecture.
Outcome:	The proposed framework outperforms baseline methods in producing informative and reliable articles.

AnswerFact: Fact Checking in Product Question Answering (2020.emnlp-main)

Copied to clipboard

Wenxuan Zhang, Yang Deng, Jing Ma, Wai Lam

Question Answering

Challenge:	a product-related community question answering platform is widely employed in many E-commerce sites . however, the misinformation in the answers on those platforms poses unprecedented challenges for users to obtain reliable and truthful product information.
Approach:	They propose a large scale fact checking dataset from product question answering forums to predict the answer veracity . each answer is accompanied by its veraity label and associated evidence sentences .
Outcome:	The proposed model outperforms baselines on the question veracity prediction task.

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

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

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.

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.

UniGDD: A Unified Generative Framework for Goal-Oriented Document-Grounded Dialogue (2022.acl-short)

Copied to clipboard

Chang Gao, Wenxuan Zhang, Wai Lam

Dialogue Systems & Conversational Agents

Challenge:	Existing studies tackle the problem of error propagation by decomposing the goal-oriented document-grounded dialogue into two sub-tasks.
Approach:	They propose to unify knowledge identification and response generation into two sub-tasks by sequentially generating grounding knowledge and response.
Outcome:	The proposed framework unifies knowledge identification and response generation and models their characteristics using a prompt-connected multi-task learning strategy.

AQE: Argument Quadruplet Extraction via a Quad-Tagging Augmented Generative Approach (2023.findings-acl)

Copied to clipboard

Jia Guo, Liying Cheng, Wenxuan Zhang, Stanley Kok, Xin Li, Lidong Bing

Syntactic Text Processing

Challenge:	Argument mining involves multiple subtasks, but each one is insufficient for understanding argumentative structure and reasoning process.
Approach:	They propose a quadruplet extraction task that extracts four argumentative components . they use a generative quadragging module to augment the training of the generative framework .
Outcome:	The proposed method can extract arguments from a large-scale dataset.

PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance (2022.emnlp-main)

Copied to clipboard

Yang Deng, Wenqiang Lei, Wenxuan Zhang, Wai Lam, Tat-Seng Chua

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:	Existing studies on financial question answering systems focus on passively responding to user queries.
Approach:	They propose a new dataset to facilitate conversational question answering over hybrid contexts in finance . they propose PACIFIC to combine clarification question generation and CQA .
Outcome:	The proposed method performs multi-task learning over all sub-tasks in PACIFIC . it incorporates a simple ensemble strategy to alleviate error propagation issue .

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

Reasoning

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 .

FreeChunker: A Cross-Granularity Chunking Framework (2026.findings-acl)

Copied to clipboard

Zhang Wenxuan, Yuan-Hao Jiang, Yang Cao, Yonghe Wu

Syntactic Text Processing

Challenge:	Existing chunking paradigms rely on static boundary identification, limiting performance . Existing methods rely only on static knowledge, resulting in hallucinated content .
Approach:	They propose a Cross-Granularity Encoding Framework that treats sentences as atomic units and shifts from static chunk segmentation to flexible retrieval supporting arbitrary sentence combinations.
Outcome:	The proposed framework avoids the computational overhead required for semantic boundary detection and enhances adaptability to complex queries.

Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations (2023.acl-long)

Copied to clipboard

Yang Deng, Wenxuan Zhang, Yifei Yuan, Wai Lam

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

Challenge:	Experimental results show the superiority of a mixed-initiative framework for emotional support conversation (ESC) ESC systems are emerging to provide prompt and convenient emotional support for helpseekers, including mental health support, counseling or motivational interviewing.
Approach:	They propose a knowledge-enhanced mixed-initiative framework that retrieves actual case knowledge from a large-scale mental health knowledge graph for generating mixed-initiative responses.
Outcome:	The proposed framework retrieves actual case knowledge from a large-scale mental health knowledge graph for generating mixed-initiative responses.

Copied to clipboard

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

PsyChain: A Collaborative Chain-of-Agents Framework for Generating Personalized and Professional Counseling Dialogues (2026.findings-acl)

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

Yi Feng, Zijie Yang, Chen Zhang, Wenxuan Zhang, Dongming Zhang, Liping Jing, Jian Yu

Dialogue Systems & Conversational Agents

Challenge:	Existing psychological counseling datasets suffer from monolithic client personas, insufficient therapeutic depth, and a lack of process controllability.
Approach:	They propose a framework that evolves static counseling corpora into high-fidelity dialogues . they use a Client Profiler that pairs life scenarios with psychological personality archetypes based on client personality and stage progression .
Outcome:	The proposed framework achieves 61-91% win rates against domain-specific baselines in pairwise evaluation and the highest average score in human evaluation, indicating potential for real-world counseling.

Aspect-based Sentiment Analysis in Question Answering Forums (2021.findings-emnlp)

Copied to clipboard

Wenxuan Zhang, Yang Deng, Xin Li, Lidong Bing, Wai Lam

FACT-AUDIT: An Adaptive Multi-Agent Framework for Dynamic Fact-Checking Evaluation of Large Language Models (2025.acl-long)

Challenge:	Existing studies on aspects-based sentiment analysis focus on a single opinionated sentence.
Approach:	They propose a model to combine aspects and their sentiments for QA forums . they use cross-sentence aspect-opinion interaction modeling to align the aspect mentioned in the question and associated opinion clues in the answer.
Outcome:	The proposed model outperforms baseline models on three real-world datasets.

Copied to clipboard

Hongzhan Lin, Yang Deng, Yuxuan Gu, Wenxuan Zhang, Jing Ma, See-Kiong Ng, Tat-Seng Chua

DimA: A Parameter-efficient Fine-tuning Method with Knowledge Transfer Based on Transformer (2024.lrec-main)

Challenge:	Existing fact-checking evaluation methods rely on static datasets and classification metrics, which fail to evaluate justification production and uncover the nuanced limitations of LLMs.
Approach:	They propose a framework that adaptively and dynamically assesses LLMs’ fact-checking capabilities by incorporating justification production alongside verdict prediction.
Outcome:	Experiments show that the framework differentiates among state-of-the-art LLMs, providing valuable insights into model strengths and limitations in model-centric fact-checking analysis.

Copied to clipboard

Wenxuan Zhang, Min Huang, Zhuoyang Song, Qinghai Miao

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

Challenge:	Pre-trained language models (PLMs) have demonstrated impressive performance across various downstream tasks, but fine-tuning is computationally expensive and storage-intensive.
Approach:	They propose a parameter-efficient method called DimA which enhances the transformer architecture by increasing the dimensionality.
Outcome:	The proposed method achieves state-of-the-art results in GLUE and XSUM tasks while utilizing less than 1% of the original model’s parameters.

Copied to clipboard

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

Can Multi-agent Help Disambiguation in Multi-domain Translation? (2026.findings-acl)

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

Zhibo Man, Shaoyang Xu, Yujie Zhang, Yi Feng, Yuanmeng Chen, Yufeng Chen, Xu Jinan, Wenxuan Zhang

Machine Translation

Challenge:	Existing multi-agent systems have shown strong potential for machine translation (MT) but their performance in multidomain translation remains unsatisfactory due to cross-domain word ambiguity .
Approach:	They propose a multi-agent collaborative disambiguation framework for MDT that leverages the collaborative capabilities of LLMs for disambiguations.
Outcome:	The proposed framework improves translation performance across multiple domains and improves disambiguation accuracy.

Seeking Patterns, Not just Memorizing Procedures: Contrastive Learning for Solving Math Word Problems (2022.findings-acl)

Copied to clipboard

Zhongli Li, Wenxuan Zhang, Chao Yan, Qingyu Zhou, Chao Li, Hongzhi Liu, Yunbo Cao

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

Challenge:	Existing models memorize procedures from context and rely on shallow heuristics to solve MWPs.
Approach:	They propose a contrastive learning approach where the neural network perceives the divergence of patterns.
Outcome:	The proposed method greatly improves performance in monolingual and multilingual settings.

Copied to clipboard

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

MPO: Multilingual Safety Alignment via Reward Gap Optimization (2025.acl-long)

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

Weixiang Zhao, Yulin Hu, Yang Deng, Tongtong Wu, Wenxuan Zhang, Jiahe Guo, An Zhang, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu

Auto-Arena: Automating LLM Evaluations with Agent Peer Battles and Committee Discussions (2025.acl-long)

Challenge:	Existing preference learning methods for safety alignment are monolingual and struggle with noisy multilingual data.
Approach:	They propose a multilingual reward gaP optimization approach that leverages the well-aligned safety capabilities of the dominant language to improve safety alignment across multiple languages.
Outcome:	Extensive experiments on three LLMs, LLaMA-3.1, Gemma-2 and Qwen2.5, validate MPO’s efficacy in multilingual safety alignment without degrading general multilingual utility.

Copied to clipboard

Ruochen Zhao, Wenxuan Zhang, Yew Ken Chia, Weiwen Xu, Deli Zhao, Lidong Bing

Pardon? Evaluating Conversational Repair in Large Audio-Language Models (2026.findings-acl)

Challenge:	Large Language Models (LLMs) are evolving rapidly and require manual evaluations.
Approach:	They propose an LLM-powered framework that automates the entire evaluation process using LLM agents.
Outcome:	The proposed framework shows a 92.14% correlation with human preferences, surpassing all previous expert-annotated benchmarks without any manual efforts.

Copied to clipboard

Shuanghong Huang, Jinlei Xu, Youchao Zhou, Yanghao Zhou, Xuan Zhao, Chong Feng, Wenxuan Zhang

Do Images Speak Louder than Words? Investigating the Effect of Textual Misinformation in VLMs (2026.eacl-long)

Challenge:	Existing evaluations of large audio-language models focus on answer accuracy and robustness to acoustic perturbations, but they assume that inputs remain semantically answerable.
Approach:	They propose a repair-aware evaluation setting that explicitly distinguishes between answerable and unanswerable audio inputs.
Outcome:	The proposed evaluation setting distinguishes between answerable and unanswerable audio inputs.

Copied to clipboard

Chi Zhang, Wenxuan Ding, Jiale Liu, Mingrui Wu, Qingyun Wu, Ray Mooney

Continual Contrastive Finetuning Improves Low-Resource Relation Extraction (2023.acl-long)

Challenge:	Existing studies have shown that Vision-Language Models have robust multimodal reasoning capabilities, but their robustness against textual misinformation remains under-explored.
Approach:	They propose to use visual-question-answering (VQA) prompts to generate persuasive prompts that deliberately conflict with visual evidence to test their models.
Outcome:	The proposed framework shows that models are vulnerable to misleading prompts, and show an average performance drop of over 48.2% after only one round of persuasive conversation.

Copied to clipboard

Wenxuan Zhou, Sheng Zhang, Tristan Naumann, Muhao Chen, Hoifung Poon

Information Extraction & Text Mining

Challenge:	Relation extraction (RE) has been challenging in low-resource domains and with limited resources.
Approach:	They propose to pretrain and finetune the RE model using consistent objectives of contrastive learning.
Outcome:	The proposed method outperforms PLM-based RE classifier on two document-level RE datasets.

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.

Knowledge Boundary of Large Language Models: A Survey (2025.acl-long)

Copied to clipboard

Moxin Li, Yong Zhao, Wenxuan Zhang, Shuaiyi Li, Wenya Xie, See-Kiong Ng, Tat-Seng Chua, Yang Deng

Easy-to-Hard Learning for Information Extraction (2023.findings-acl)

Challenge:	Large language models (LLMs) store vast amount of knowledge in their parameters, but they still have limitations in the memorization and utilization of certain knowledge.
Approach:	They propose a comprehensive definition of the LLM knowledge boundary and introduce a formalized taxonomy categorizing knowledge into four distinct types.
Outcome:	The proposed definition of the LLM knowledge boundary and taxonomy categorizes knowledge into four distinct types . aims to offer a comprehensive overview, facilitate access to key issues, and inspire further advancements in LLM research.

Copied to clipboard

Chang Gao, Wenxuan Zhang, Wai Lam, Lidong Bing

Information Extraction & Text Mining

Challenge:	Existing models for information extraction (IE) use a one-stage learning strategy to extract the target structure from unstructured text data.
Approach:	They propose a unified easy-to-hard learning framework that mimics the human learning process by breaking down the learning process into multiple stages.
Outcome:	The proposed framework enables the model to acquire general IE task knowledge and improve its generalization ability on 13 out of 17 datasets.

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

Copied to clipboard

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

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

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

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.

Disentangling Language and Culture for Evaluating Multilingual Large Language Models (2025.acl-long)

Copied to clipboard

Jiahao Ying, Wei Tang, Yiran Zhao, Yixin Cao, Yu Rong, Wenxuan Zhang

Intra-/Inter-Interaction Network with Latent Interaction Modeling for Multi-turn Response Selection (2020.coling-main)

Challenge:	Extensive evaluations of large language models (LLMs) are conducted on a wide range of models, revealing a notable cultural-linguistic synergy phenomenon, where models exhibit better performance when questions are culturally aligned with the language.
Approach:	They propose a Dual Evaluation Framework to comprehensively assess the multilingual capabilities of large language models by decomposing evaluation along dimensions of linguistic medium and cultural context.
Outcome:	The proposed framework allows for a nuanced analysis of LLMs’ ability to process questions within both native and cross-cultural contexts cross-lingually.

Copied to clipboard

Yang Deng, Wenxuan Zhang, Wai Lam

Challenge:	Existing methods for multi-turn response selection are not practical as the turns of conversations vary.
Approach:	They propose to use latent interaction modeling to model multi-level interactions between utterance and response.
Outcome:	The proposed method outperforms state-of-the-art methods on three multi-turn response selection benchmark datasets.

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

Multimodality

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.

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

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

SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe (2026.acl-long)

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

Yuxin Xiao, Shujian Zhang, Marzyeh Ghassemi, Wenxuan Zhou

SeaExam and SeaBench: Benchmarking LLMs with Local Multilingual Questions in Southeast Asia (2025.findings-naacl)

Challenge:	Efforts to improve instruction tuning often focus on higher-quality supervised fine-tuning datasets, typically requiring data filtering with proprietary LLMs or human annotation.
Approach:	They propose a Mixup-based recipe that elevates LLM instruction tuning without relying on well-curated datasets.
Outcome:	The proposed model improves instruction-following and healthcare-specific tasks with consistent improvements across LLM families and SFT datasets.

Copied to clipboard

Chaoqun Liu, Wenxuan Zhang, Jiahao Ying, Mahani Aljunied, Anh Tuan Luu, Lidong Bing

Zero-Shot Text Classification via Self-Supervised Tuning (2023.findings-acl)

Challenge:	Large Language Models (LLMs) have shown remarkable performance across various English benchmarks, including both human exam datasets such as MMLU and instruction-following datasets.
Approach:	They introduce two new benchmarks to evaluate the capabilities of Large Language Models in Southeast Asian (SEA) application scenarios.
Outcome:	The proposed benchmarks show that they can discern LLM performance on SEA language tasks compared to their translated benchmarks.

Copied to clipboard

Chaoqun Liu, Wenxuan Zhang, Guizhen Chen, Xiaobao Wu, Anh Tuan Luu, Chip Hong Chang, Lidong Bing

Low-Resource NLP

Challenge:	Existing solutions to zero-shot text classification use pre-trained language models or large-scale annotated data.
Approach:	They propose a self-supervised learning paradigm to solve zero-shot text classification tasks by tuning the language models with unlabeled data.
Outcome:	The proposed model outperforms the state-of-the-art models on 7 out of 10 tasks and is less sensitive to prompt design.

Answering Product-related Questions with Heterogeneous Information (2020.aacl-main)

Copied to clipboard

Wenxuan Zhang, Qian Yu, Wai Lam

Bidirectional Generative Framework for Cross-domain Aspect-based Sentiment Analysis (2023.acl-long)

Challenge:	Existing product question answering methods only consider a single information source such as user reviews and/or require large amounts of labeled data.
Approach:	They propose a framework to exploit heterogeneous information including natural language text and attribute-value pairs from two information sources of the concerned product, namely product details and user reviews.
Outcome:	The proposed framework achieves superior performance over state-of-the-art models on a real-world dataset.

Copied to clipboard

Yue Deng, Wenxuan Zhang, Sinno Jialin Pan, Lidong Bing

Multi-hop Inference for Question-driven Summarization (2020.emnlp-main)

Challenge:	Aspect-based sentiment analysis (ABSA) is a task of analyzing people's sentiments at the aspect level.
Approach:	They propose a unified bidirectional generative framework to tackle cross-domain ABSA tasks . the framework trains a model in both text-to-label and label-totext directions .
Outcome:	The proposed framework trains a model in both label-to-label and label- to-text directions to learn domain-agnostic features.

Copied to clipboard

Yang Deng, Wenxuan Zhang, Wai Lam

Summarization

Challenge:	Existing methods for summarizing source document for non-factoid questions are lacking in factoidic QA.
Approach:	They propose a question-driven abstractive summarization method that incorporates multi-hop reasoning into question-based summarizing.
Outcome:	The proposed method outperforms state-of-the-art methods on two non-factoid QA datasets.

Cross-lingual Aspect-based Sentiment Analysis with Aspect Term Code-Switching (2021.emnlp-main)

Copied to clipboard

Wenxuan Zhang, Ruidan He, Haiyun Peng, Lidong Bing, Wai Lam

WPO: Enhancing RLHF with Weighted Preference Optimization (2024.emnlp-main)

Challenge:	Existing studies on Aspect-based sentiment analysis (ABSA) focus on English texts, but handling it in resource-poor languages remains a challenge.
Approach:	They propose an unsupervised cross-lingual transfer method for the Aspect-based sentiment analysis task . they propose an aspect code-switching mechanism to augment training data with code-linked bilingual sentences .
Outcome:	The proposed method preserves task-specific knowledge in the target language.

Copied to clipboard

Wenxuan Zhou, Ravi Agrawal, Shujian Zhang, Sathish Reddy Indurthi, Sanqiang Zhao, Kaiqiang Song, Silei Xu, Chenguang Zhu

Self-regulation: Employing a Generative Adversarial Network to Improve Event Detection (P18-1)

Challenge:	Off-policy preference optimization suffers from a distributional gap between the policy used for data collection and the target policy, leading to suboptimal optimization.
Approach:	They propose a method to simulate on-policy learning with off-police preference data.
Outcome:	The proposed method outperforms Direct Preference Optimization (DPO) by up to 5.6% on Alpaca Eval 2 and MT-bench.

Copied to clipboard

Yu Hong, Wenxuan Zhou, Jingli Zhang, Guodong Zhou, Qiaoming Zhu

Robustness in NLP

Challenge:	Recent studies show that neural networks can be used for event detection but can be contaminated by spurious features.
Approach:	They propose a self-regulated learning approach by utilizing a generative adversarial network to generate spurious features.
Outcome:	The proposed method is highly effective and adaptable on the ACE 2005 and TAC-KBP 2015 corpora.

Exploiting BERT for End-to-End Aspect-based Sentiment Analysis (D19-55)

Copied to clipboard

Xin Li, Lidong Bing, Wenxuan Zhang, Wai Lam

Is Translation All You Need? A Study on Solving Multilingual Tasks with Large Language Models (2025.naacl-long)

Challenge:	Existing studies on ABSA use a sequence tagging problem to extract aspect-specific opinion words from the sentence given the aspect.
Approach:	They build a series of simple yet insightful neural baselines to deal with E2E-ABSA task using contextualized embeddings from pre-trained language models.
Outcome:	The proposed architecture outperforms state-of-the-art models even with a simple linear classification layer.

Copied to clipboard

Chaoqun Liu, Wenxuan Zhang, Yiran Zhao, Anh Tuan Luu, Lidong Bing

Zero-to-Strong Generalization: Eliciting Strong Capabilities of Large Language Models Iteratively without Gold Labels (2025.coling-main)

Challenge:	Large language models (LLMs) have demonstrated multilingual capabilities, yet they are mostly English-centric due to the imbalanced training corpora.
Approach:	They extend the evaluation to real-world user queries and non-English-centric LLMs . they show that translation into English can boost LLM performance on NLP tasks .
Outcome:	The proposed evaluation extends to user queries and non-English-centric LLMs . it shows that translation into English can boost performance on NLP tasks, but not universally optimal .

Copied to clipboard

Chaoqun Liu, Qin Chao, Wenxuan Zhang, Xiaobao Wu, Boyang Li, Anh Tuan Luu, Lidong Bing

T-REG: Preference Optimization with Token-Level Reward Regularization (2025.acl-long)

Challenge:	Pre-trained language models have demonstrated remarkable performance through supervised fine-tuning or in-context learning using gold labels.
Approach:	They propose a new paradigm termed zero-to-strong generalization that prompts LLMs to annotate unlabeled data and retain high-quality labels by filtering.
Outcome:	The proposed framework outperforms pre-trained language models on extensive classification and reasoning tasks on multiple model sizes.

Copied to clipboard

Wenxuan Zhou, Shujian Zhang, Lingxiao Zhao, Tao Meng

Language of Thought Shapes Output Diversity in Large Language Models (2026.acl-long)

Challenge:	Reinforcement learning from human feedback (RLHF) is a dominant approach for large language models to follow instructions and produce meaningful alignment.
Approach:	They propose a method that leverages human feedback to optimize large language models . they propose to use sequence-level and token-level rewards to optimize preference .
Outcome:	The proposed method outperforms baseline methods on Alpaca Eval 2 and Arena-Hard benchmarks.

Copied to clipboard

Shaoyang Xu, Wenxuan Zhang

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

Challenge:	Using multiple thinking languages, we show that controlling the language used during model thinking provides a novel and structural source of output diversity.
Approach:	They propose to control the language used during model thinking to provide a novel source of output diversity.
Outcome:	The proposed methods show that controlling the language used during model thinking provides a novel and structural source of output diversity.

Copied to clipboard

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

Towards Generalizable and Robust Text-to-SQL Parsing (2022.findings-emnlp)

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

Chang Gao, Bowen Li, Wenxuan Zhang, Wai Lam, Binhua Li, Fei Huang, Luo Si, Yongbin Li

Programming Languages in NLP

Challenge:	Text-to-SQL parsers must be generalizable and robust against input perturbations.
Approach:	They propose a novel framework to learn text-to-SQL parsing in stages to improve parser's ability to acquire general SQL knowledge instead of capturing spurious patterns.
Outcome:	The proposed framework achieves state-of-the-art performance on the Spider, SParC, and CoSQL datasets.

Sentiment Analysis in the Era of Large Language Models: A Reality Check (2024.findings-naacl)

Copied to clipboard

Wenxuan Zhang, Yue Deng, Bing Liu, Sinno Pan, Lidong Bing

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

Challenge:	Sentiment analysis (SA) has been a long-standing research area in natural language processing.
Approach:	They propose a benchmark to evaluate LLMs' SA abilities and propose 'sentiEval' benchmark to be used for a more comprehensive evaluation.
Outcome:	The proposed benchmark outperforms small language models on 26 datasets on 13 tasks and compared them with LLMs trained on domain-specific datasets.

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

Towards Generative Aspect-Based Sentiment Analysis (2021.acl-short)

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

Wenxuan Zhang, Xin Li, Yang Deng, Lidong Bing, Wai Lam

Sharpness-Aware Minimization with Dynamic Reweighting (2022.findings-emnlp)

Challenge:	Existing work on Aspect-based sentiment analysis ignores the rich label semantics of ABSA.
Approach:	They propose to tackle various ABSA tasks in a unified generative framework . they propose to use annotation-style and extraction-style modeling to enable training .
Outcome:	The proposed framework achieves state-of-the-art on four ABSA tasks across multiple benchmark datasets.

Copied to clipboard

Wenxuan Zhou, Fangyu Liu, Huan Zhang, Muhao Chen

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

Challenge:	Deep neural networks are often overparameterized and can overfit training data.
Approach:	They propose an adversarial weight minimization algorithm that conducts adversarials and finds a common adversaria per-batch.
Outcome:	The proposed algorithm finds a common adversarial weight perturbation per-batch.

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.

Pruning General Large Language Models into Customized Expert Models (2025.findings-acl)

Copied to clipboard

Yiran Zhao, Guizhen Chen, Kenji Kawaguchi, Lidong Bing, Wenxuan Zhang

JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning (2025.findings-acl)

Challenge:	Large language models (LLMs) require significant computational resources to maintain their general capabilities.
Approach:	They propose a Custom Pruning method to prune a large general model into a smaller lightweight expert model, positioned along the "language", "domain" and "task" dimensions.
Outcome:	The proposed method outperforms existing pruning methods and achieves minimal loss in both expert and general capabilities across models from different model families and sizes.

Copied to clipboard

Chang Gao, Wenxuan Zhang, Guizhen Chen, Wai Lam

Hiring Now: A Skill-Aware Multi-Attention Model for Job Posting Generation (2020.acl-main)

Challenge:	Existing text-to-text methods struggle with issues such as generalization, robustness, and controllability due to their lack of explicit task structures.
Approach:	They propose a structure-to-structure approach that uses JSON structures to represent tasks.
Outcome:	The proposed method outperforms TextTuning in terms of performance, robustness, and controllability across different scenarios.

Copied to clipboard

Liting Liu, Jie Liu, Wenzheng Zhang, Ziming Chi, Wenxuan Shi, Yalou Huang

Challenge:	Creating job requirements is a crucial step in the recruiting process, but it is difficult to specify the level of education, experience, relevant skills per the job description.
Approach:	They propose a conditional text generation task to generate job requirements based on job descriptions . they use a hierarchical decoder to label the job description with multiple skills . a skill knowledge graph is constructed to capture the global prior knowledge about skills based upon the model .
Outcome:	The proposed method is evaluated on real-world job posting data.

On-the-fly Denoising for Data Augmentation in Natural Language Understanding (2024.findings-eacl)

Copied to clipboard

Tianqing Fang, Wenxuan Zhou, Fangyu Liu, Hongming Zhang, Yangqiu Song, Muhao Chen

Low-Resource NLP

Challenge:	Existing methods to improve data augmentation performance may introduce noisy data that impairs training.
Approach:	They propose an on-the-fly denoising technique that learns from soft augmented labels provided by an organic teacher model trained on the cleaner original dataset.
Outcome:	The proposed method improves on text classification and question-answering tasks on general augmentation techniques and prevents overfitting on noisy labels.

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

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

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

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

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

Multimodality

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.

Cultivating Gaming Sense for Yourself: Making VLMs Gaming Experts (2025.acl-long)

Copied to clipboard

Wenxuan Lu, Jiangyang He, Zhanqiu Zhang, Steven Y. Guo, Tianning Zang

Multimodality

Challenge:	Recent efforts leverage Vision Language Models (VLMs) as direct controllers, often pausing the game to analyze screens and plan action through language reasoning.
Approach:	They propose a paradigm shift in gameplay agent design that uses Vision Language Models as a developer instead of direct control.
Outcome:	The proposed framework achieves fluent gameplay in diverse genres, including ACT, FPS, and Flappy Bird, setting a new benchmark for game-playing agents.

Aspect Sentiment Quad Prediction as Paraphrase Generation (2021.emnlp-main)

Copied to clipboard

Wenxuan Zhang, Yang Deng, Xin Li, Yifei Yuan, Lidong Bing, Wai Lam

Context-faithful Prompting for Large Language Models (2023.findings-emnlp)

Challenge:	Existing studies focus on predicting the four elements in one shot, instead of predicting them all.
Approach:	They propose a task to jointly detect all sentiment elements in quads for a given opinionated sentence.
Outcome:	The proposed method can generate the semantics of the sentiment elements in the natural language form.

Copied to clipboard

Wenxuan Zhou, Sheng Zhang, Hoifung Poon, Muhao Chen

Do LLMs Really Know What They Don’t Know? Internal States Mainly Reflect Knowledge Recall Rather Than Truthfulness (2026.findings-acl)

Challenge:	Large language models encode parametric knowledge about world facts but overly rely on it can cause incorrect predictions in context-sensitive NLP tasks.
Approach:	They propose to use opinion-based prompts and counterfactual demonstrations to improve LLM faithfulness to contexts.
Outcome:	The proposed methods improve faithfulness to contexts using opinion-based prompts and counterfactual demonstrations.

Copied to clipboard

Chi Seng Cheang, Hou Pong Chan, Wenxuan Zhang, Yang Deng

Reasoning

Challenge:	Recent work suggests that large language models (LLMs) produce hallucinated and factually correct outputs.
Approach:	They propose a taxonomy categorizing hallucinations into Unassociated Hallucination (UH) and Associated Hallucinian (AH) they propose to use internal signals to distinguish hallucinos from factual errors .
Outcome:	The proposed taxonomy categorizes hallucinations into Unassociated Hallucination (UH) and Associated Hallucinications (AHs) based on the proposed taxonomic, the authors show that hidden states reflect whether the model is recalling parametric knowledge rather than the truthfulness of the output itself.

DR-Arena: an Automated Evaluation Framework for Deep Research Agents (2026.acl-long)

Copied to clipboard

Yiwen Gao, Ruochen Zhao, Yang Deng, Wenxuan Zhang

Challenge:	Existing benchmarks for evaluating deep research capabilities rely on static datasets.
Approach:	They propose a fully automated evaluation framework that pushes DR agents to their capability limits through dynamic investigation.
Outcome:	DR-Arena achieves a Spearman correlation of 0.94 with the LMSYS Search Arena leaderboard.

On the Multi-turn Instruction Following for Conversational Web Agents (2024.acl-long)

Copied to clipboard

Yang Deng, Xuan Zhang, Wenxuan Zhang, Yifei Yuan, See-Kiong Ng, Tat-Seng Chua