PhD student · UCAS / AMSS

Chen Han 韩晨

Research interests

  • Large language models
  • Agentic RL
  • Multi-agent systems, harness

I am a PhD student in a joint program between the School of Advanced Interdisciplinary Sciences (SAIS) and the Academy of Mathematics and Systems Science (AMSS), Chinese Academy of Sciences. I expect to graduate in 2028.Outside of research, I am also a Top 500 Tank player in Overwatch — 汉什么森#5664

News

  • 文章“基于大语言模型的政府招商引资知识图谱构建及应用”被中文核心期刊《系统工程理论与实践》录用。
  • One paper titled “Beyond Factual Knowledge: Benchmarking and Learning Step-Level Procedural Rule Reasoning in Large Language Models” has been accepted by EMNLP 2026.
  • 入选小红书 ACE 人才计划,并加入小红书实习。
  • One paper titled “Beyond Detection: Exploring Evidence-based Multi-Agent Debate for Misinformation Intervention and Persuasion” has been accepted by AAAI 2026 (Oral). First-time Oral Link, 50:00
  • 入选中国科协青年科技人才培育工程博士生专项计划。
  • 荣获中国科学院大学比亚迪奖学金。
  • One paper titled “Debate-to-Detect: Reformulating Misinformation Detection as a Real-World Debate with Large Language Models” has been accepted by EMNLP 2025.
  • One paper titled “DocPolicyKG: A Lightweight LLM-Based Framework for Knowledge Graph Construction from Chinese Policy Documents” has been accepted by CIKM 2025.

Selected research

All publications
AAAI 2026Oral

Beyond Detection: Exploring Evidence-based Multi-Agent Debate for Misinformation Intervention and Persuasion [Oral ]

Chen Han, Yijia Ma, Jin Tan, Wenzhen Zheng, Xijin Tang

Evidence-grounded multi-agent debate for misinformation intervention and persuasion.

EMNLP 2025

Debate-to-Detect: Reformulating Misinformation Detection as a Real-World Debate with Large Language Models

Chen Han, Wenzhen Zheng, Xijin Tang

Reframing misinformation detection as a real-world debate between language model agents.

CIKM 2025

DocPolicyKG: A Lightweight LLM-Based Framework for Knowledge Graph Construction from Chinese Policy Documents

Chen Han, Yuanyuan Li, Xijin Tang

A lightweight framework for policy knowledge graphs and retrieval-augmented question answering.

Full publication list11 publications

2026

  1. Yu, B., Cao, P., Han, C., Zhou, C., Zhang, Z., Xie, Z., Teng, W., Liao, X., Zhao, J., & Liu, K. (2026). Beyond factual knowledge: Benchmarking and learning step-level procedural rule reasoning in large language models. Findings of the Association for Computational Linguistics: EMNLP 2026. Paper ↗

  2. Han, C., Ma, Y., Tan, J., & Tang, X. (2026). Beyond detection: Exploring evidence-based multi-agent debate for misinformation intervention and persuasion. Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2026 Oral). Paper ↗

2025

  1. Han, C., & Tang, X. (2025). An agent-based simulation framework for misinformation susceptibility test with LLMs: Insights from psychological factors. In Knowledge and Systems Sciences (pp. 289–302). Springer, Singapore. Paper ↗

  2. Han, C., Zheng, W., & Tang, X. (2025). Debate-to-detect: Reformulating misinformation detection as a real-world debate with large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (pp. 15114–15129). Association for Computational Linguistics. Paper ↗

  3. Han, C., Li, Y., & Tang, X. (2025). DocPolicyKG: A lightweight LLM-based framework for knowledge graph construction from Chinese policy documents. In Proceedings of the 34th ACM International Conference on Information and Knowledge Management (pp. 4753–4757). Paper ↗

  4. 韩晨, 唐锡晋. ACM TechNews中虚假信息报道发展进程研究—基于科技新闻的文本挖掘分析. 系统科学与数学, 2025. Paper ↗ (Han, C., & Tang, X. (2025). Research on the development of disinformation reports in ACM TechNews: A text mining analysis based on technology news. Journal of Systems Science and Mathematical Sciences.)

2023 & earlier

  1. Han, C., Wu, C., & Wei, L. (2023). The impact of the disclosure characteristics of the application material on the successful listing of companies on China’s Science and Technology Innovation Board. Journal of Behavioral and Experimental Finance, 37, 100733.

  2. Chen, N., Han, C., & Wei, L. (2023). The impact of subjective and objective inconsistencies in scientific and technological innovation attributes on the listing of enterprises on the Science and Technology Innovation Board. Procedia Computer Science, 221, 493–500.

  3. Wei, L., Deng, Y., Huang, J., Han, C., & Jing, Z. (2022). Identification and analysis of financial technology risk factors based on textual risk disclosures. Journal of Theoretical and Applied Electronic Commerce Research, 17(2), 590–612.

  4. Wei, L., Han, C., & Yao, Y. (2022). The bias analysis of oil and gas companies’ credit ratings based on textual risk disclosures. Energies, 15(7), 2390.

  5. Han, C., Liang, D., Huang, J., & Wei, L. (2022). The impact of disclosure characteristics of the registration statement for Science and Technology Innovation Board on the IPO underpricing. Procedia Computer Science, 199, 238–245.

Experience

Xiaohongshu 小红书

ACE Top Talent

Data Engineer Agent R&D, LLM posttrain

Beijing

Ant Group 蚂蚁集团

LLM Algorithm Intern

Post-training & Evaluation · Beijing

Source Code Capital 源码资本

Algorithm Intern

Beijing

Education

University of Chinese Academy of Sciences

Ph.D. · Computer Science and Technology

School of Advanced Interdisciplinary Sciences

Advisor: Xijin Tang ↗

Chinese Academy of Sciences

M.S. · Computer Science and Technology

Academy of Mathematics and Systems Science

Advisor: Xijin Tang ↗

Central University of Finance and Economics

B.S. · Big Data Management and Application

Advisor: Lu Wei ↗

Honors & awards

Earlier awards
  • Merit Student and Outstanding Student Leader, UCAS(中国科学院大学三好学生、优秀学生干部)
  • Outstanding Graduate and Outstanding Undergraduate Thesis, CUFE (中央财经大学优秀毕业生、优秀毕业论文)
  • National Undergraduate Innovation and Entrepreneurship Training Program(国家级大学生创新创业训练项目优秀结项)
  • Second Prize, MathorCup University Mathematical Modeling Challenge(MathorCup高校数学建模挑战赛大数据竞赛二等奖)
  • First Prize, Mathematical Contest in Modeling (MCM/ICM) (美国大学生数学建模竞赛一等奖)
  • 中央财经大学三好学生

Academic service

Conference reviewer

  • AAAI
  • AISI
  • CIKM
  • ICLR
  • AAMAS
  • WWW
  • KSS