🤵🏻 About Me

I am a researcher with research interests in artificial intelligence and machine learning. I obtained my Ph.D. degree from Department of Computer Science and Technology in Nanjing University in June 2022, where I was very fortunate to be advised by professor Yu-Feng Li (李宇峰). Currently, I am an Assistant Professor in School of Intelligence Science and Technology, Nanjing University (Suzhou Campus). I am also a member of LAMDA Group (机器学习与数据挖掘研究所), which is led by professor Zhi-Hua Zhou (周志华).

📖 Research

📢: 招收对神经符号学习、大模型、Agent、具身智能感兴趣，有较好编程基础的学生。此外，长期招收全职或者兼职的实习生，本科、硕士、博士均可申请。 如有兴趣请给我发送邮件(guolz@nju.edu.cn)，附上你的简历，简要介绍你的研究兴趣和研究经历。更多内容见：招生说明。

📢: 每周面向本科生开放办公室交流，可以交流课程、学业、生活等各方面的问题，提前约时间即可~

The long-term research goal of our team is to enhance the reasoning and planning capabilities of AI agent in both digital and physical worlds, contributing to the advancement of artificial general intelligence (AGI). Our core approach centers on neuro-symbolic learning, which bridges data-driven machine learning with knowledge-driven symbolic reasoning, often regarded as the hallmark of third-generation AI. The neural component provides grounding in perception and physical interaction, while the symbolic component augments reasoning and planning.

Recently, our research has primarily focused on overcoming the limitations of reasoning and planning capabilities in Large Language Models (LLMs) and Multi-Modal Large Language Models (MLLMs), and on building AI agents capable of thinking, reasoning, planning, and acting in complex, dynamic, multi-modal environments. Our work spans the following directions:

• LLM / MLLM Reasoning: Constructing high-quality reasoning data; post-training algorithms (SFT, RL); tool-augmented reasoning, and agentic reasoning frameworks. These techniques are applied to mathematical reasoning, medical reasoning, legal reasoning, remote sensing interpretation, and other related domains.
• Agent in Digital World: Building autonomous agents that operate in digital environments, navigating web interfaces, manipulating software tools, writing code, and orchestrating complex tool-use workflows.
• Agent in Physical World: Building embodied agents and game agents that perceive, reason, and act in physical or simulated 3D environments. Key challenges include long-horizon task planning, spatial reasoning, multi-agent coordination, and sim-to-real transfer.
• LLMs for Open Problems: Exploring the frontier of using LLMs as reasoning engines for unsolved problems in mathematics and natural sciences, including automated theorem proving, symbolic regression, scientific discovery, and augmenting human researchers with AI-driven conjecture and verification loops.

🎉 News

• 2026.01: Three Papers are accepted by ICLR 2026!
• 2026.01: Congratulations on winning the Best Student Paper Award at the AAAI 2026 Trust Agent Workshop!
• 2025.11: One Paper is accepted by AAAI 2026!
• 2025.11: One Paper is accepted by AAAI 2026 Trust Agent Workshop!
• 2025.10: Congratulations to the three students (Hao-Zhe Tan, Zi-Jian Cheng, Song-Lin Lv) who received the National Scholarship!
• 2025.10: One Paper is accepted by MATH-AI Workshop @ NeurIPS 2025!
• 2025.09: One Paper is accepted by NeurIPS 2025!
• 2025.08: One Paper is accepted by EMNLP 2025!
• 2025.07: One Paper is accepted by AI4Math Workshop in ICML 2025!
• 2025.05: We have released TabFSBench for evaluation tabular data learning in open environments.
• 2025.05: Three Papers are accepted by ICML 2025!
• 2025.04: One Paper is accepted by IJCAI 2025 Survey Track!
• 2025.03: One Paper is accepted by SCI-FM Workshop @ ICLR 2025!
• 2025.01: One Paper is accepted by ICLR 2025!
• 2024.12: We have released ChinaTravel benchmark for travel planning.
• 2024.12: One Paper is accepted by AAAI 2025!
• 2024.10: We have released Awesome Neuro-Symbolic Learning in the Era of Foundation Models.
• 2024.05: One Paper is accepted by FCS!
• 2024.05: One Paper is accepted by KDD 2024!
• 2024.04: One Paper is accepted by ICML 2024!
• 2024.01: One Paper is accepted by ICLR 2024!
• 2023.12: CFP: We are organizing a "Robust Machine Learning in Open Environments" workshop in PAKDD 2024, Taipei, May 7, 2024
• 2023.12: One Paper is accepted by IEEE TVCG!
• 2023.11: One paper is accepted by Science China Information Science!
• 2023.05: We open-sourced a LLM for Chinese Legal domain LawGPT.⛏️ 🛠️
• 2022.07: We open-sourced the semi-supervised learning toolkit LAMDA-SSL.⛏️ 🛠️