Peng Li is an Associate Research Professor at the Institute for AI Industry Research (AIR), Tsinghua University. He received his bachelor's and doctoral degrees from the Department of Computer Science and Technology, Tsinghua University. His research interests include Large Language Models (LLMs), AI Agents, AI for Mathematics (AI4Math), and AI Scientist. He has published more than 100 papers in major international conferences and journals in artificial intelligence.
He has received several research awards, including the ACL 2023 Outstanding Paper Award, the ICAIS 2025 Responsible AI Research Award, and the First Prize of the Qian Weichang Chinese Information Processing Science and Technology Award from the Chinese Information Processing Society of China (CIPS). He has led major research projects, including a key task under the National Science and Technology Innovation - Major Program and a General Program project of the National Natural Science Foundation of China. He has also served as an Area Chair for top-tier international conferences such as ICLR, ACL, EMNLP, and NAACL. His research has been deployed in large-scale production systems at WeChat and Baidu.
Homepage
http://www.lpeng.net
Education
· Ph.D. in Computer Science, Tsinghua University, 2009-2015
· B.Eng. in Computer Science, Tsinghua University, 2005-2009
Professional Experience
· Associate Research Professor, Institute for AI Industry Research (AIR), Tsinghua University, 2022-Present
· Principal Researcher & Team Leader, Pattern Recognition Center, WeChat, Tencent, 2018-2022
· Senior Research Engineer, Institute of Deep Learning (IDL), Baidu, 2015-2017
Research Interests
Large Language Models (LLMs), AI Agents, AI for Mathematics (AI4Math), AI Scientist
Selected Publications
1. AI Mathematician: Towards Fully Automated Frontier Mathematical Research.
Yuanhang Liu, Yanxing Huang, Yanqiao Wang, Peng Li, Yang Liu. arXiv.
Introduces the AI Mathematician (AIM) framework, exploring how AI can participate in frontier mathematical research and advancing AI4Math from benchmark solving toward mathematical discovery.
System: https://github.com/TheoryFoundry
Blog: https://ai-mathematician.net
2. Pessimistic Verification for Open Ended Math Questions
Yanxing Huang, Zihan Tang, Zejin Lin, Peng Li, Yang Liu. ICML 2026.
Introduces the Pessimistic Verification framework for detecting reasoning errors in open-ended mathematical problems, improving the reliability of AI-based verification for mathematical proofs and reasoning.
3. AIGS: Generating Science from AI-Powered Automated Falsification
Zijun Liu, Kaiming Liu, Yiqi Zhu, Xuanyu Lei, Zonghan Yang, Zhenhe Zhang, Peng Li, Yang Liu. Preprint.
Introduces the AI-Generated Science (AIGS) framework, which places automated falsification at the core of scientific discovery and enables AI systems to generate, test, and refine scientific hypotheses.
4. Position: Towards Unified Alignment Between Agents, Humans, and Environment
Zonghan Yang, An Liu, Zijun Liu, Kaiming Liu, Fangzhou Xiong, Yile Wang, Zeyuan Yang, Qingyuan Hu, Xinrui Chen, Zhenhe Zhang, Fuwen Luo, Zhicheng Guo, Peng Li, Yang Liu. ICML 2024.
Proposes the Unified Alignment of Agents, Humans, and Environment (UA2) paradigm, extending AI alignment beyond human preferences to incorporate environmental constraints and practical considerations such as time and cost.
5. A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration
Zijun Liu, Yanzhe Zhang, Peng Li, Yang Liu, Diyi Yang. COLM 2024.
Introduces DyLAN, a dynamic multi-agent collaboration framework that adaptively optimizes collaboration patterns according to task requirements and improves collaborative reasoning on complex tasks.
Honors & Awards
1. ICAIS 2025 Responsible AI Research Award (2025)
2. ACL 2023 Outstanding Paper Award (2023)
3. CIKM 2021 Best Resource Paper Nomination (2021)
4. First Prize of the Qian Weichang Chinese Information Processing Science and Technology Award (2020)
