自然语言处理，大模型理论与应用

（1）Ph.D., University of Reading, England（雷丁大学，计算机系，2009年）

（2）M.Sc., Nanjing University, China（南京大学，计算机系，2003年）

（3）B.Sc., Nanjing University, China（南京大学，数学系，2000年）

（1）2009年，剑桥大学，博士后

（2）2010年，东南大学计算机学院，副教授

（3）2016年，东南大学计算机学院，教授

主持完成国家自然科学基金项目2项、省部级项目3项（教育部博士点新教师基金、教育部回国留学人员基金、江苏自然科学基金面上项目），联合主持完成国家自然科学基金港澳及海外学者合作项目1项。

1. 国家自然科学基金 基于贝叶斯框架的面向多源数据的事件概率化表示学习研究 2022-2025

2. 国家自然科学基金 面向热门争议话题的基于社交网络文本与结构的层次观点挖掘研究 2018-2021

3. 国家自然科学基金 面向社交网络的意见群体检测研究 2016-2017

4. 国家自然科学基金 基于简标注和弱监督学习的开放的信息抽取研究 2012-2014

5. 江苏省自然基金 面向多源异构网络文本的事件抽取研究 2017-2019

Selected Publications（详细见google scholar）：

大模型方向：

2025

1. Ziyang Ma, Qingyue Yuan, Zhenglin Wang, Deyu Zhou. Large Language Models Have Intrinsic Meta-Cognition, but Need a Good Lens.  EMNLP, pp 3460–3477, 2025. 

2. Yilong Lai, Jialong Wu, Zhenglin Wang, Deyu Zhou. AdaRewriter: Unleashing the Power of Prompting-based Conversational Query Reformulation via Test-Time Adaptation. EMNLP, pp 3889–3905, 2025. 

3. Jialong Wu, Zhenglin Wang, Linhai Zhang, Yilong Lai, Yulan He, Deyu Zhou. Scope: Optimizing key-value cache compression in long-context generation. ACL, pp 10775-10790, 2025.

4. Congzhi Zhang, Linhai Zhang, Jialong Wu, Yulan He and Deyu Zhou*. Causal Prompting: Debiasing Large Language Model Prompting based on Front-Door Adjustment. AAAI, pp 25842-25850, 2025

5. Yanyue Zhang, Yulan He and Deyu Zhou*. Rehearse With User: Personalized Opinion Summarization via Role-Playing based on Large Language Models. ACL, pp 15194–15211, 2025.

6. Linhai Zhang, Jialong Wu, Deyu Zhou* and Yulan He. PROPER: A Progressive Learning Framework for Personalized Large Language Models with Group-Level Adaptation. ACL, pp 16399–16411, 2025.

7. Linhai Zhang, Ziyang Gao, Deyu Zhou* and Yulan He. Explainable Depression Detection in Clinical Interviews with Personalized Retrieval-Augmented Generation. ACL, pp 9927–9944, 2025.

8. Jialong Wu, Wenbiao Yin, Yong Jiang, Zhenglin Wang, Zekun Xi, Linhai Zhang, Yulan He, Deyu Zhou. WebWalker: Benchmarking LLMs in Web Traversal. ACL, pp 10290–10305, 2025.

9. Zhenglin Wang, Jialong Wu, Yilong Lai, Congzhi Zhang, Deyu Zhou. Seed: Accelerating reasoning tree construction via scheduled speculative decoding, COLING, pp 4920-4937, 2025.

10. Yanyue Zhang, Pengfei Li, Yilong Lai, Yulan He and Deyu Zhou*. LASS: A Novel and Economical Data Augmentation Framework Based on Language Models for Debiasing Opinion Summarization. COLING, pp 6169-6183, 2025.

自然语言处理：

2025

1. Xueguan Zhao, Wenpeng Lu, Chaoqun Zheng, Weiyu Zhang, Jiasheng Si, Deyu Zhou. Plan Dynamically, Express Rhetorically: A Debate-Driven Rhetorical Framework for Argumentative Writing. EMNLP, pp 9562-9584, 2025.

2. Jiasheng Si, Haiyang Zhu, Yibo Zhao, Weiyu Zhang, Tianyi Wang, Wenpeng Lu, Deyu Zhou. Scene Generalization for Biomedical Fact Verification via Hierarchical Mixture of Experts. Information Sciences, pp 122527, 2025

3. Yanyue Zhang, Yilong Lai, Zhenglin Wang, Deyu Zhou. DimSum: Disentangling Representation with Automatically Generated Multi-Category Summary Templates for Fine-grained Opinion Summarization. Expert Systems with Applications, 2025

4. Xin Zhang, Qiyu Wei, Yingjie Zhu, Linhai Zhang, Deyu Zhou, Sophia Ananiadou. SynGraph: A Dynamic Graph-LLM Synthesis Framework for Sparse Streaming User Sentiment Modeling. ACL, pp 16338–16356, 2025

5. Linhai Zhang, Xin Zhang, Deyu Zhou. Relevant emotion ranking with topic-enhanced emotion transition. Knowledge-Based Systems, pp 113015, 2025.

6. Yilong Lai, Jialong Wu, Congzhi Zhang, Haowen Sun, Deyu Zhou. AdaCQR: Enhancing query reformulation for conversational search via sparse and dense retrieval alignment. COLING, pp 7698-7720, 2025.

7. Wenjie Wei, Yanyue Zhang, Jinyan Li, Panfei Liu and Deyu Zhou*. Cross-Domain Fake News Detection based on Dual-Granularity Adversarial Training. COLING, pp 9407-9417, 2025.

2024

1. Jiasheng Si, Yibo Zhao, Yingjie Zhu, Haiyang Zhu, Wenpeng Lu and Deyu Zhou*. CHECKWHY: Causal Fact Verification via Argument Structure. ACL, pp 15636–15659, 2024. (CCF A类，杰出论文奖)

2. Congzhi Zhang, Linhai Zhang and Deyu Zhou*. Causal Walk: Debiasing Multi-Hop Fact Verification with Front-Door Adjustment. AAAI, pp 19533- 19541, 2024. 

3. Yuanyuan Xu, Linhai Zhang and Deyu Zhou*. TECA: A Two-stage Approach with Controllable Attention Soft Prompt for Few-shot Nested Named Entity Recognition. COLING, pp 15698–15710, 2024. 

4. Yanyue Zhang, Yilong Lai, Zhenglin Wang, Pengfei Li, Deyu Zhou* and Yulan He. Opinions Are Not Always Positive: Debiasing Opinion Summarization with Model-Specific and Model-Agnostic Methods. COLING, pp 12496–12513, 2024. 

5. Jialong Wu, Linhai Zhang, Deyu Zhou, Guoqiang Xu. DINER: Debiasing Aspect-based Sentiment Analysis with Multi-variable. ACL, pp  3504–3518, 2024

6. Linhai Zhang, Jialong Wu, Deyu Zhou, Guoqiang Xu. Star: Constraint lora with dynamic active learning for data-efficient fine-tuning of large language models. ACL, pp 3519–3532, 2024.

2023

1. Xin Zhang, Linhai Zhang and Deyu Zhou*. Sentiment Analysis on Streaming User Reviews via Dual-Channel Dynamic Graph Neural Network. EMNLP, pp 7208–7220, 2023. 

2. Yingjie Zhu, Jiasheng Si, Yibo Zhao, Haiyang Zhu, Deyu Zhou* and Yulan He. EXPLAIN, EDIT, GENERATE: Rationale-Sensitive Counterfactual Data Augmentation for Multi-hop Fact Verification. EMNLP, pp 13377–13392, 2023. 

3. Yuanyuan Xu, Zeng Yang, Linhai Zhang, Deyu Zhou*, Tiandeng Wu and Rong Zhou. Focusing, Bridging and Prompting for Few-shot Nested Named Entity Recognition. ACL, pp 2621–2637, 2023.

4. Yanzheng Xiang, Qian-Wen Zhang, Xu Zhang, Zejie Liu, Yunbo Cao and Deyu Zhou*. G3R: A Graph-Guided Generate-and-Rerank Framework for Complex and Cross-domain Text-to-SQL Generation. ACL, pp 338–352, 2023.

5. Boyu Wang, Linhai Zhang, Deyu Zhou*, Yi Cao and Jiandong Ding. Neural Topic Modeling based on Cycle Adversarial Training and Contrastive Learning. ACL, pp 9720–9731, 2023.

6. Yanyue Zhang and Deyu Zhou*. Disentangling Text Representation with Counter-Template for Unsupervised Opinion Summarization. ACL, pp 6344–6357, 2023.

7. Jiasheng Si, Yingjie Zhu and Deyu Zhou*. Exploring Faithful Rationale for Multi-hop Fact Verification via Salience-aware Graph Learning. AAAI, pp 13573–13581, 2023.

2022

1. Zejie Liu, Xiaoyu Hu, Deyu Zhou*, Lin Li, Xu Zhang and Yanzheng Xiang. Code Generation From Flowcharts With Texts: a Benchmark Dataset and an Approach. Findings of EMNLP, pp 6069–6077, 2022.

2. Zeng Yang, Linhai Zhang and Deyu Zhou*. SEE-few: Seed, Expand and Entail for Few-shot Named Entity Recognition. COLING, pp 2540–2550, 2022. 

3. Linhai Zhang and Deyu Zhou*. Temporal Knowledge Graph Completion With Approximated Gaussian Process Embedding. COLING, pp 4697–4706, 2022. 

4. Xu Zhang, Zejie Liu, Yanzheng Xiang and Deyu Zhou*. Complicate Then Simplify: A Novel Way to Explore Pre-trained Models for Text Classification. COLING, pp 1136–1145, 2022. 

5. Linhai Zhang, Xuemeng Hu, Boyu Wang, Deyu Zhou*, Qian-Wen Zhang and Yunbo Cao. Pre-training and Fine-tuning Neural Topic Model: a Simple yet Effective Approach to Incorporating External Knowledge. ACL, pp 5980-5989, 2022. 

6. Tao Wang, Linhai Zhang, Chenchen Ye, Junxi Liu and Deyu Zhou*. A Novel Framework Based on Medical Concept Driven Attention for Explainable Medical Code Prediction via External Knowledge. Findings of ACL, pp 1407-1416, 2022.

2021

1. Chenchen Ye, Linhai Zhang, Yulan He, Deyu Zhou* and Jie Wu. Beyond text: Incorporating Metadata and Label Structure for Multi-label Document Classification using Heterogeneous Graphs. EMNLP, pp 3162-3171, 2021.

2. Deyu Zhou*, Jianan Wang, Linhai Zhang and Yulan He. Implicit Sentiment Analysis with Event-centered Text Representation. EMNLP, pp 6884-6893, 2021. 

3. Deyu Zhou*, Yanzheng Xiang, Linhai Zhang, Chenchen Ye, Qian-Wen Zhang and Yunbo Cao. A Divide-And-Conquer Approach for Multi-label Multi-hop Relation Detection in Knowledge Base Question Answering. EMNLP, pp 4798-4808, 2021.

4. Linhai Zhang, Deyu Zhou*, Chao Lin and Yulan He. A Multi-label Multi-hop Relation Detection Model Based on Relation-aware Sequence Generation. EMNLP, pp 4713-4719, 2021.

5. Jiasheng Si, Deyu Zhou*, Tongzhe Li, Xingyu Shi and Yulan He. Topic-aware Evidence Reasoning and Stance-aware Aggregation for Fact Verification. ACL, pp 1612–1622, 2021. 

6. Lixing Zhu, Gabriele Pergola, Lin Gui, Deyu Zhou and Yulan He. Topic-driven and Knowledge-aware Transformer for Dialogue Emotion Detection. ACL, pp 1571–1582, 2021. 

7. Linhai Zhang, Deyu Zhou*, Yulan He and Zeng Yang. MERL: Multimodal Event Representation Learning in Heterogeneous Embedding Spaces. AAAI, pp. 14420–14427, 2021. 

8. Deyu Zhou*, Meng Zhang, Linhai Zhang and Yulan He. A Neural Group-wise Sentiment Analysis Model with Data Sparsity Awareness. AAAI, pp. 14594–14601, 2021.

软件工程：

1. Zexu Lin, Xu Zhang, Xiaoyu Hu, Pengfei Li, Deyu Zhou, Darong Lai. Hierarchical Mixture-of model for unified code search. Information Fusion, 2026.

2. Xu Zhang, Zexu Lin, Xiaoyu Hu, Jianlei Wang, Wenpeng Lu and Deyu Zhou*. SECON: Maintaining Semantic Consistency in Data Augmentation for Code Search. TOIS, pp 1-26, 2025.

3. Xiaoyu Hu, Xu Zhang, Zexu Lin and Deyu Zhou*. De-Redundancy then Rethink: Enhancing Code Summarization with a Novel Pipeline Framework. COLING, pp 13722–13733, 2024. 

4. Xu Zhang, Xiaoyu Hu, Deyu Zhou. CORES: COde REpresentation Summarization for Code Search. IEEE Transactions on Consumer Electronics, 2024.2 

1, 面向社交媒体文本的无监督的事件抽取和分类方法

2, 一种基于纠错输出编码的无需消歧的无监督词性标注方法

3, 一种端到端的基于上下文的知识库问答方法及装置

4, 面向生物医学文本的基于主题注意机制的多标签分类方法

5, 一种面向机场应急预案文本的实体与关系联合抽取方法

6, 一种面向小规模知识图谱的链接预测方法

7, 一种面向新闻文本的基于深度聚类模型的事件线抽取方法

8, 一种面向稀疏数据的个性化情感分析方法与装置

9, 面向新闻文本基于层级状态神经网络的社会情绪排序方法

10, 一种基于社交媒体内容和结构的用户实时观点检测方法

11, 一种引入类信息的文本生成图像方法及装置

12, 一种引入胶囊网络的文本生成图像方法及装置

13, 一种基于深度交互推理模型的事实检测方法

14, 基于预训练语言模型的抑郁症访谈对话生成方法

15, 一种基于显式及隐式特征解耦的NL2SQL方法

16, 基于双层多门控专家混合模型的多任务NL2SQL方法

17, 一种基于分层特征解耦的NL2SQL方法和系统 

18, 基于三阶段的小样本嵌套命名实体识别方法和系统

19, 基于反模版解耦的无监督意见摘要生成方法及系统

20, 基于对抗性框架的无监督意见摘要生成方法及系统

21, 基于双层多门控专家混合模型的多任务NL2SQL方法

ACL 2024年 杰出论文奖。