Papers by Huimin Wang

Full-Step-DPO: Self-Supervised Preference Optimization with Step-wise Rewards for Mathematical Reasoning (2025.findings-acl)

Challenge:	Existing text-based recommendation frameworks that use pretrained language models (PLMs) can improve performance on text-related tasks.
Approach:	They propose a unified local- and global-attention Transformer encoder to better model two-level contexts of user history.
Outcome:	The proposed framework improves on three text-based recommendation tasks.

Copied to clipboard

Huimin Xu, Xin Mao, Feng-Lin Li, Xiaobao Wu, Wang Chen, Wei Zhang, Anh Tuan Luu

Reasoning

Challenge:	Existing approaches to improve long-chain mathematical reasoning focus on the first erroneous step, but ignore all other steps and rely heavily on external signals.
Approach:	They propose a DPO framework that leverages step-wise rewards from the entire reasoning chain instead of optimizing only the first erroneous step.
Outcome:	The proposed framework improves on in-domain and out-of-domain mathematical reasoning benchmarks.

Domain Adaptation for Question Answering via Question Classification (2022.coling-1)

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Ziyi Kou, Lanyu Shang, Dong Wang

QA Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation (2022.emnlp-main)

Challenge:	Question answering systems often experience performance deterioration upon user-generated questions.
Approach:	They propose a question classification framework to help QA domains adapt to different domains.
Outcome:	The proposed framework improves on state-of-the-art datasets against multiple datasets.

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Bernhard Kratzwald, Stefan Feuerriegel, Dong Wang

Low-Resource NLP

Challenge:	Question answering models often suffer from performance deterioration upon deployment .
Approach:	They propose a self-supervised framework called QADA for QA domain adaptation . they propose to augment training QA samples with hidden space augmentation .
Outcome:	The proposed framework improves on multiple target datasets over state-of-the-art methods.

ACBQ: Adaptive Cross-Block Quantization of Large Language Models (2026.acl-long)

Copied to clipboard

Hailing Wang, Jianglin Lu, Yitian Zhang, Huimin Zeng, Yun Fu

Guaranteeing Knowledge Integration with Joint Decoding for Retrieval-Augmented Generation (2026.acl-long)

Challenge:	Existing methods for post-training quantization struggle to support weight–activation joint quantization and extreme low-bit weight quantization.
Approach:	They propose a framework that addresses weight–activation joint quantization and extreme weight quantization.
Outcome:	The proposed framework achieves superior performance under both W4A4 and highly aggressive W2 settings while incurring negligible additional computational overhead.

Copied to clipboard

Zhengyi Zhao, Shubo Zhang, Zezhong Wang, Yuxi Zhang, Huimin Wang, Yutian Zhao, Yefeng Zheng, Binyang Li, Kam-Fai Wong, Xian Wu

Reasoning

Challenge:	Retrieval-Augmented Generation (RAG) provides access to external knowledge, but current research focuses on retrieval quality and 'integration bottleneck' .
Approach:	They propose a framework that explicitly decouples reasoning from evidence integration by generating an 'Inner-Answer' and a 'Refer-Aswer" they propose 'a joint decoding mechanism that dynamically fuses the logical coherence of the Inner-Andswer with the factual precision of the Refer-Adswer at the token level'
Outcome:	The proposed framework improves accuracy by 12.1% and reduces hallucinations by 16.3% on five QA benchmarks.

Retrieval Augmented Fact Verification by Synthesizing Contrastive Arguments (2024.acl-long)

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Lanyu Shang, Yifan Liu, Yang Zhang, Dong Wang

Can LLMs Replace Clinical Doctors? Exploring Bias in Disease Diagnosis by Large Language Models (2024.findings-emnlp)

Challenge:	Existing methods to verify claim credibility rely on embedded knowledge or unreliable context.
Approach:	They propose retrieval augmented fact verification through the synthesis of contrasting arguments (RAFTS) they use an embedding model to identify informative demonstrations and in-context prompts to generate the prediction and explanation.
Outcome:	The proposed method outperforms existing methods with smaller LLMs or unreliable contexts.

Copied to clipboard

Yutian Zhao, Huimin Wang, Yuqi Liu, Wu Suhuang, Xian Wu, Yefeng Zheng

Reaction Miner: An Integrated System for Chemical Reaction Extraction from Textual Data (2023.emnlp-demo)

Challenge:	a new study examines the bias of disease prediction in large language models . the model biases are prevalent across gender, age range and disease judgment behaviors .
Approach:	They propose a prompt-based approach to alleviate the bias in disease prediction with LLMs.
Outcome:	The proposed model alleviates the observed bias in disease prediction with LLMs.

Copied to clipboard

Ming Zhong, Siru Ouyang, Yizhu Jiao, Priyanka Kargupta, Leo Luo, Yanzhen Shen, Bobby Zhou, Xianrui Zhong, Xuan Liu, Hongxiang Li, Jinfeng Xiao, Minhao Jiang, Vivian Hu, Xuan Wang, Heng Ji, Martin Burke, Huimin Zhao, Jiawei Han

Information Extraction & Text Mining

Challenge:	Reaction Miner is a system designed to extract chemical reactions from raw scientific PDFs.
Approach:	They propose a system that extracts chemical reactions directly from raw scientific PDFs.
Outcome:	The proposed system can extract chemical reactions from raw scientific PDFs.

A Layered Debating Multi-Agent System for Similar Disease Diagnosis (2025.naacl-short)

Copied to clipboard

Yutian Zhao, Huimin Wang, Yefeng Zheng, Xian Wu

Challenge:	Traditional classification, contrastive learning, and large language models fail to detect subtle clues necessary for differentiation.
Approach:	They propose a framework that leverages Large Language Models to achieve accurate disease diagnosis . they structure patient information and integrate extensive medical knowledge to guide the analysis .
Outcome:	The proposed framework aims to identify subtle differences between similar diseases . the proposed framework can be used in clinical practice to improve accuracy .

Revealing the Seen, Imagining the Beyond: A Survey of Image-Grounded Chain-of-Thought Reasoning in Multimodal LLMs (2026.acl-long)

Copied to clipboard

Qihua Dong, Yitian Zhang, Huimin Zeng, Yizhou Wang, Jianglin Lu, Kuo Yang, Yun Fu

Zero- and Few-Shot Event Detection via Prompt-Based Meta Learning (2023.acl-long)

Challenge:	Recent advances in Multimodal Large Language Models (MLLMs) have shifted visual reasoning from tool-calling to end-to-end perceptionreasoning.
Approach:	They synthesize the emerging paradigm of Image-Grounded Chain-of-Thought (IG-CoT) they propose a method-centric taxonomy covering prompting, supervised fine-tuning, and reinforcement learning .
Outcome:	The proposed model is based on a method-centric taxonomy and benchmarks.

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Mengfei Lan, Heng Ji, Dong Wang

Learning Efficient Dialogue Policy from Demonstrations through Shaping (2020.acl-main)

Challenge:	Existing methods for event detection often fail to detect unseen or rare events due to the lack of domain knowledge.
Approach:	They propose a meta learning-based framework for zero-shot event detection that uses a prompt-based prompt and a trigger-aware soft verbalizer to efficiently project output to unseen tasks.
Outcome:	The proposed framework performs state-of-the-art in zero-shot and few-shot scenarios on benchmark datasets FewEvent and MAVEN.

Copied to clipboard

Huimin Wang, Baolin Peng, Kam-Fai Wong

Dialogue Systems & Conversational Agents

Challenge:	Using reinforcement learning to learn dialogue policy requires a large volume of interactions with users.
Approach:	They propose a task-oriented dialogue agent that efficiently learns dialogue policy from demonstrations . they use an imitation model to distill knowledge from demonstration and reward shaping .
Outcome:	The proposed agent efficiently learns dialogue policy from demonstrations through policy shaping and reward shaping.

MemeReaCon: Probing Contextual Meme Understanding in Large Vision-Language Models (2025.emnlp-main)

Copied to clipboard

Zhengyi Zhao, Shubo Zhang, Yuxi Zhang, Yanxi Zhao, Yifan Zhang, Zezhong Wang, Huimin Wang, Yutian Zhao, Bin Liang, Yefeng Zheng, Binyang Li, Kam-Fai Wong, Xian Wu

Negation Triplet Extraction with Syntactic Dependency and Semantic Consistency (2024.lrec-main)

Challenge:	Current approaches focus on isolated meme analysis, either for harmful content detection or standalone interpretation, overlooking a fundamental challenge: the same meme can express different intents depending on its conversational context.
Approach:	They propose a benchmark to evaluate how large vision language models understand memes in their original context.
Outcome:	The proposed benchmark evaluates how large vision language models understand meme intent in their original context.

Copied to clipboard

Yuchen Shi, Deqing Yang, Jingping Liu, Yanghua Xiao, Zongyu Wang, Huimin Xu

SCOPE: Compress Mathematical Reasoning Steps for Efficient Automated Process Annotation (2025.findings-acl)

Challenge:	Negation understanding is crucial to many downstream tasks such as sentiment analysis, question answering, Web search and natural language inference.
Approach:	They propose a novel negation triplet extraction task which aims to extract negation subject along with negation cue and scope.
Outcome:	The proposed model is based on a generative pretrained language model with a multi-task learning framework and achieves the best performance compared to baselines.

Copied to clipboard

Huimin Xu, Xin Mao, Feng-Lin Li, Xiaobao Wu, Wang Chen, Wei Zhang, Anh Tuan Luu

Reasoning

Challenge:	Existing process annotation approaches are computationally expensive.
Approach:	They propose a compression-based approach that transforms reasoning steps into code and normalizes them through Abstract Syntax Tree.
Outcome:	The proposed method outperforms existing methods on Best-of-N strategy and ProcessBench.

imapScore: Medical Fact Evaluation Made Easy (2024.findings-acl)

Copied to clipboard

Huimin Wang, Yutian Zhao, Xian Wu, Yefeng Zheng

Don’t Forget Your Reward Values: Language Model Alignment via Value-based Calibration (2024.emnlp-main)

Challenge:	Automated evaluation of natural language generation tasks fails to focus on medical QA because of the diversity in medical terminology.
Approach:	They propose a new data structure, imap, to capture key information in questions and answers.
Outcome:	The proposed model outperforms state-of-the-art metrics in correlation with human scores.

Copied to clipboard

Xin Mao, Feng-Lin Li, Huimin Xu, Wei Zhang, Wang Chen, Anh Tuan Luu

RareSyn: Health Record Synthesis for Rare Disease Diagnosis (2025.emnlp-main)

Challenge:	Existing methods for generating large language models have been criticized for their complexity and instability.
Approach:	They propose a value-based calibration method to better align Large Language Models with human preferences.
Outcome:	The proposed method surpasses existing methods on AI assistant and summarization datasets, providing impressive generalizability, robustness, and diversity in different settings.

Copied to clipboard

Huimin Wang, Yutian Zhao, Yefeng Zheng, Xian Wu

Introducing Compiler Semantics into Large Language Models as Programming Language Translators: A Case Study of C to x86 Assembly (2024.findings-emnlp)

Challenge:	RareSyn is a data synthesis approach to augment and de-identify EHRs with a focus on rare diseases.
Approach:	They propose a data synthesis approach to augment and de-identify EHRs with a focus on rare diseases.
Outcome:	The proposed model augments and de-identifies EHRs with a focus on rare diseases.

Copied to clipboard

Shuoming Zhang, Jiacheng Zhao, Chunwei Xia, Zheng Wang, Yunji Chen, Huimin Cui

Fair Federated Learning with Biased Vision-Language Models (2024.findings-acl)

Challenge:	Large Language Models (LLMs) can be used to translate high-level programming languages to machine instructions.
Approach:	They propose two methods to solve a problem known as neural compilation by using a 13B model with a behavioral accuracy of over 91%.
Outcome:	The proposed approach outperforms the larger model by over 50% and achieves a behavioral accuracy of over 91% while outperforming the GPT-4 Turbo model.

Copied to clipboard

Huimin Zeng, Zhenrui Yue, Yang Zhang, Lanyu Shang, Dong Wang

TaiChi: Improving the Robustness of NLP Models by Seeking Common Ground While Reserving Differences (2024.lrec-main)

Challenge:	Existing literature ignores the inherent group unfairness within CLIP and its ethical implications on FL applications.
Approach:	They propose a fairness-aware adaptation framework for CLIP in federated learning . they propose to leverage biased pre-trained VLMs to build fair FL frameworks .
Outcome:	The proposed framework addresses unique bias in FL, triggered by data heterogeneity . it trains a fair FL model with fairness-aware deep visual prompting (DVP) Extensive results on human face attribute recognition (FAR) applications show it outperforms state-of-the-art FL models .

Copied to clipboard

Huimin Chen, Chengyu Wang, Yanhao Wang, Cen Chen, Yinggui Wang

Robustness in NLP

Challenge:	Pre-trained Language Models are vulnerable to adversarial examples that introduce human-imperceptible perturbations to clean examples to deceive the models.
Approach:	They propose a Siamese network-based approach to teach adversarial models to focus on similarities . they propose combining two sub-networks sharing the same structure but trained on clean and adversarials .
Outcome:	The proposed approach reduces the differences between clean and adversarial samples and focuses more on similarities.

Chem-FINESE: Validating Fine-Grained Few-shot Entity Extraction through Text Reconstruction (2024.findings-eacl)

Copied to clipboard

Qingyun Wang, Zixuan Zhang, Hongxiang Li, Xuan Liu, Jiawei Han, Huimin Zhao, Heng Ji

T2: An Adaptive Test-Time Scaling Strategy for Contextual Question Answering (2025.emnlp-main)

Challenge:	Existing frameworks for fine-grained few-shot entity extraction are difficult to implement in the chemical domain due to the information overload of scientific papers.
Approach:	They propose a sequence-to-sequence based few-shot entity extraction approach . it uses a seq2seq entity extractor and a self-validation module to reconstruct original input sentence .
Outcome:	The proposed framework achieves 8.26% and 6.84% performance gains on two datasets.

Copied to clipboard

Zhengyi Zhao, Shubo Zhang, Zezhong Wang, Huimin Wang, Yutian Zhao, Bin Liang, Yefeng Zheng, Binyang Li, Kam-Fai Wong, Xian Wu

Question Answering

Challenge:	Existing efficient test-time scaling methods introduce budget constraints or early stop mechanisms to avoid overthinking for straightforward questions but add human bias to the reasoning process.
Approach:	They propose a framework that dynamically adapts reasoning depth based on question complexity.
Outcome:	Experimental results show that the proposed framework achieves higher accuracy than baseline methods and reduces computational overhead by up to 25.2%.

Distorted or Fabricated? A Survey on Hallucination in Video LLMs (2026.findings-acl)

Copied to clipboard

Yiyang Huang, Yitian Zhang, Yizhou Wang, Mingyuan Zhang, Liang Shi, Huimin Zeng, Yun Fu

Self-DC: When to Reason and When to Act? Self Divide-and-Conquer for Compositional Unknown Questions (2025.naacl-long)

Challenge:	Despite significant advances in video-language modeling, hallucinations remain a persistent challenge in video large language models.
Approach:	They present a systematic taxonomy that categorizes hallucinations into two core types: dynamic distortion and content fabrication.
Outcome:	The proposed taxonomy categorizes hallucinations into two core types: dynamic distortion and content fabrication.

Copied to clipboard

Hongru Wang, Boyang Xue, Baohang Zhou, Tianhua Zhang, Cunxiang Wang, Huimin Wang, Guanhua Chen, Kam-Fai Wong

CoAD: Automatic Diagnosis through Symptom and Disease Collaborative Generation (2023.acl-long)

Challenge:	Existing studies focus on leveraging internal knowledge of Large Language Models (LLMs) to answer known questions.
Approach:	They propose a framework that allows LLMs to choose between internal and external knowledge . they use a dataset to analyze compositional questions that are composed of unknown sub-questions .
Outcome:	The proposed framework can achieve comparable or even better performance with much fewer external calls compared with several strong baselines.

Copied to clipboard

Huimin Wang, Wai Chung Kwan, Kam-Fai Wong, Yefeng Zheng

Challenge:	Automated diagnosis (AD) is a critical application of AI in healthcare . despite its simplicity and superior performance, a decline in disease diagnosis accuracy is observed .
Approach:	They propose a new collaborative disease and symptom generation framework to improve automatic diagnosis.
Outcome:	The Transformer-based method achieves an average 2.3% improvement over previous state-of-the-art methods . it can be used to query patients about their symptoms and health concerns .

Train Once, Deploy Anywhere: Matryoshka Representation Learning for Multimodal Recommendation (2024.findings-emnlp)

Copied to clipboard

Yueqi Wang, Zhenrui Yue, Huimin Zeng, Dong Wang, Julian McAuley

Monte Carlo Tree Search Based Prompt Autogeneration for Jailbreak Attacks against LLMs (2025.coling-main)

Challenge:	integrating rich multimodal knowledge into recommender systems remains a challenge . despite performance improvements, different recommendation scenarios often require varying granularities.
Approach:	They propose a framework that captures item features at different granularities and learns informative representations for efficient recommendation across multiple dimensions.
Outcome:	The proposed framework achieves superior performance over state-of-the-art models on multiple benchmark datasets.

Copied to clipboard

Suhuang Wu, Huimin Wang, Yutian Zhao, Xian Wu, Yefeng Zheng, Wei Li, Hui Li, Rongrong Ji

JoTR: A Joint Transformer and Reinforcement Learning Framework for Dialogue Policy Learning (2024.lrec-main)

Challenge:	Jailbreak attacks craft specific prompts or append adversarial suffixes to prompts, thereby inducing language models to generate harmful or unethical content and bypassing the model’s safety guardrails.
Approach:	They propose a Monte Carlo Tree Search (MCTS) based Prompt Auto-generation (MPA) method to generate adversarial suffixes for valid jailbreak attacks.
Outcome:	The proposed method outperforms existing methods on open-source and closed-source models and shows that it can generate harmful responses.

Copied to clipboard

Wai-Chung Kwan, Huimin Wang, Hongru Wang, Zezhong Wang, Bin Liang, Xian Wu, Yefeng Zheng, Kam-Fai Wong

Dialogue Systems & Conversational Agents

Challenge:	Dialogue policy learning (DPL) aims to determine an abstract representation (also known as action) to guide what the response should be.
Approach:	They propose a joint Transformer-based model that generates a token-grained policy that allows more dynamic dialogue action generation without the need for predefined action candidates.
Outcome:	The proposed model outperforms existing models showing improvements of 9% and 13% in success rate and 34% and 37% in diversity of dialogue actions across two benchmark dialogue modeling tasks.

Multi-Agent-as-Judge: Aligning LLM-Agent-Based Automated Evaluation with Multi-Dimensional Human Evaluation (2026.acl-long)

Copied to clipboard

Jiaju Chen, Yuxuan Lu, Xiaojie Wang, Huimin Zeng, Jing Huang, Jiri Gesi, Ying Xu, Bingsheng Yao, Dakuo Wang

Challenge:	Existing "LLM-as-a-judge" evaluation frameworks are limited by persona descriptions and are not generalizable to other tasks.
Approach:	They propose a framework that can automatically construct multiple evaluator personas with distinct dimensions from relevant text documents and instantiate LLM agents with the persona.
Outcome:	The proposed framework can believably simulate human evaluators . it extracts stakeholders' diverse perspectives from the provided research papers and constructs personas for the agents .

Evidence-Driven Retrieval Augmented Response Generation for Online Misinformation (2024.naacl-long)

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Yimeng Lu, Lanyu Shang, Yang Zhang, Dong Wang

Text Generation

Challenge:	Existing methods to generate counter-misinformation responses are often trained end-to-end without external knowledge, resulting in subpar text quality and excessively repetitive responses.
Approach:	They propose retrieval augmented response generation for online misinformation (RARG) that collects supporting evidence and generates counter-misinformation responses via reinforcement learning from human feedback.
Outcome:	The proposed method outperforms baselines with extensive experiments with in- and cross-domain datasets and consistently generates high-quality counter-misinformation responses.

A Collaborative Multi-agent Reinforcement Learning Framework for Dialog Action Decomposition (2021.emnlp-main)

Copied to clipboard

Huimin Wang, Kam-Fai Wong

Dialogue Systems & Conversational Agents

Challenge:	Currently, most reinforcement learning methods for dialog policy learning train a centralized agent that selects a predefined joint action concatenating domain name, intent type, and slot name.
Approach:	They propose a hierarchical multi-agent framework in which each part of the action is led by a different agent and a joint optimization process that makes agents can exchange their policy information.
Outcome:	The proposed framework reduces labor costs for action templates and decreases the size of the action space for each agent.

Scaling up Open Tagging from Tens to Thousands: Comprehension Empowered Attribute Value Extraction from Product Title (P19-1)

Copied to clipboard

Huimin Xu, Wenting Wang, Xin Mao, Xinyu Jiang, Man Lan

Information Extraction & Text Mining

Challenge:	Existing models treat each attribute as an entity type and build one set of NER tags for each of them, leading to scalability issues.
Approach:	They propose to regard attribute as a query and adopt only one global set of BIO tags for any attributes to reduce the burden of attribute tag or model explosion.
Outcome:	The proposed model outperforms state-of-the-art models and generates promising results for 8,906 attributes.

MetaAdapt: Domain Adaptive Few-Shot Misinformation Detection via Meta Learning (2023.acl-long)

Copied to clipboard

Zhenrui Yue, Huimin Zeng, Yang Zhang, Lanyu Shang, Dong Wang

Bridging the Temporal Gap in Multimodal LLMs: Deeply Stacking Temporal Tokens for Audio-Visual Speech Recognition (2026.findings-acl)

Challenge:	Existing methods for misinformation detection are limited by data scarcity . existing methods fail to detect early-stage misinformation on emerging topics .
Approach:	They propose a meta learning based approach for domain adaptive few-shot misinformation detection that leverages limited target examples to provide feedback and guide the knowledge transfer from the source to the target domain.
Outcome:	The proposed method improves performance on real-world datasets with reduced parameters.

Copied to clipboard

Liyong Wang, Junliang Xing, Tianyu Hu, Jianfei Jiang, Jihuai Zhao, Huimin Ma

SoulChat: Improving LLMs’ Empathy, Listening, and Comfort Abilities through Fine-tuning with Multi-turn Empathy Conversations (2023.findings-emnlp)

Challenge:	Existing audio-visual speech recognition systems suffer from a temporal gap . visual speech patterns captured from lip movements provide complementary information that remains inherently robust to acoustic noise.
Approach:	They propose a framework that deeply stacks temporal tokens across both encoding and decoding stages to bridge this temporal gap.
Outcome:	The proposed framework outperforms existing supervised, self-supervised, and LLM-based methods by 6.1% on LRS2 and 7.8% on LLS3.

Copied to clipboard

Yirong Chen, Xiaofen Xing, Jingkai Lin, Huimin Zheng, Zhenyu Wang, Qi Liu, Xiangmin Xu

MKeCL: Medical Knowledge-Enhanced Contrastive Learning for Few-shot Disease Diagnosis (2024.lrec-main)

Challenge:	Large language models (LLMs) are used in psychological counseling to provide universal advice.
Approach:	They constructed a multi-turn empathetic conversation dataset with 2 million samples . they found that the model's empathy ability is enhanced when finetuning .
Outcome:	Experiments show that large language models can be finetuned to provide empathy . but, when applied to mental health or emotional support conversation, there are three main issues .

Copied to clipboard

Yutian Zhao, Huimin Wang, Xian Wu, Yefeng Zheng