Welcome!
Hi, nice to meet you! I’m Xu Jundong, a first-year PhD Student at the National University of Singapore (NUS) CTIC lab, working under the supervision of Prof. Mong-Li Lee and Prof. Wynne Hsu. I am also mentored by Dr. Hao Fei. Previously, I earned a Master of Computing from NUS and a BSc from University College London (UCL).
My work focuses on enhancing the symbolic reasoning capabilities of computational systems. I aim to develop methods that support systematic and reliable inference, helping to bridge the gap between data-driven approaches and structured logical reasoning.
I am especially interested in interdisciplinary collaborations and welcome discussions with researchers working on topics related to logic, reasoning, and neuro-symbolic frameworks. Check out our Neuro-Symbolic Model project website. Please feel free to reach out via email to talk about research!
🔥 News
- 2026.06: 🎉🎉 The paper EvoArena is released on Arxiv and rank as #1 Hugging Face Daily Paper!
- 2026.04: 🎉🎉 Honored to be invited to Tsinghua University to deliver a talk on Towards Reliable Logical Reasoning in Large Foundation Models.
- 2026.02: 🎉🎉 The paper MuSLR received the Best Paper Award at the AAAI 2026 Logical Reasoning Workshop. We are truly grateful for the recognition!
- 2026.01: 🎉🎉 The paper LogicReward is accepted to ICLR 2026. See you in Rio de Janeiro!
- 2025.12: 🎉🎉 The paper LogicReward is released on Arxiv.
- 2025.09: 🎉🎉 The paper MuSLR has been accepted to NeurIPS 2025. See you in San Diego!
- 2025.06: 🎉🎉 The paper Aristotle has been selected as an oral presentation (8% of the accepted paper).
- 2025.05: 🎉🎉 The paper Aristotle has been accepted to ACL 2025 main conference.
- 2025.04: 🎉🎉 I got a PhD offer from NUS!
- 2025.01: 🎉🎉 I started working as a research assistant at NUS CTIC.
- 2025.01: 🎉🎉 I officially received my master’s degree.
- 2024.12: 🎉🎉 Our new paper Aristotle is released on Arxiv.
- 2024.10: 🎉🎉 One paper was accepted at the ACM MM Workshop on Multi-modal Misinformation Governance in the Era of Foundation Models.
- 2024.05: 🎉🎉 The paper SymbCoT has been accepted to the ACL 2024 main conference.
📝 Publications
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
Jundong Xu*, Qingchuan Li*, Jiaying Wu, Yihuai Lan, Shuyue Stella Li, Huichi Zhou, Bowen Jiang, Lei Wang, Jun Wang, Anh Tuan Luu, Caiming Xiong, Hae Won Park, Bryan Hooi, Zhiyuan Hu
- EvoArena evaluates LLM agents in environments where terminal workflows, software repositories, and user preferences evolve over time. EvoMem augments agent memory with patch histories that preserve what changed, why it changed, and when old behavior still matters.
LogicReward: Incentivizing LLM Reasoning via Step-Wise Logical Supervision
Jundong Xu, Hao Fei*, Huichi Zhou, Xin Quan, Qijun Huang, Shengqiong Wu, William Yang Wang, Mong-Li Lee, Wynne Hsu.
- LogicReward is a reward function that evaluates unstructured natural language reasoning and provides step-level, symbolically guided rewards.
MuSLR: Multimodal Symbolic Logical Reasoning
Jundong Xu, Hao Fei*, Yuhui Zhang, Liangming Pan, Qijun Huang, Qian Liu, Preslav Nakov, Min-Yen Kan, William Yang Wang, Mong-Li Lee, Wynne Hsu.
- We present the first multimodal symbolic reasoning benchmark grounded in formal logical rules, and propose a modular framework, LogiCAM, to enhance VLMs’ MuSLR capabilities.
Jundong Xu, Hao Fei*, Meng Luo, Qian Liu, Liangming Pan, William Yang Wang, Preslav Nakov, Mong-Li Lee, Wynne Hsu.
- This paper integrates a logic-complete decompose–search–resolve framework to strengthen LLM reasoning with symbolic logic.
Faithful Logical Reasoning via Symbolic Chain-of-Thought
Jundong Xu, Hao Fei*, Liangming Pan, Qian Liu, Mong-Li Lee, Wynne Hsu.
- This paper combines symbolic logical rules with the reasoning chain to improve logical reasoning in LLMs.
📖 Educations
- 2025.08 - Present, PhD in Computer Science, National University of Singapore
- 2023.01 - 2025.01, Master of Computing, National University of Singapore
- 2019.09 - 2022.07, BSc Information Management, University College London
🏅 Honors and Rewards
- AAAI 2026 Symbolic and Logical Reasoning Workshop Best Paper Award.
- NUS Research Achievement Award.
- Hugging Face Daily Paper #1 (EvoArena).
💬 Invited Talks
- 2026.04, Invited to Tsinghua University to deliver a talk on Towards Reliable Logical Reasoning in Large Foundation Models.
- 2026.01, AAAI 2026 Symbolic and Logical Reasoning Workshop Oral Presentation to introduce paper MuSLR.
- 2025.11, Invited to NICE (NLP Academic Exchange) to introduce NeurIPS 2025 paper MuSLR. | [video]
- 2025.05, Invited to NICE (NLP Academic Exchange) to introduce ACL 2025 paper Aristotle. | [video]
- 2024.06, Invited to NICE (NLP Academic Exchange) to introduce ACL 2024 paper SymbCoT. | [video]
💻 Academic Services
- Reviewer: ICLR, NeurIPS, ACL, EMNLP, ACM MM, Neurocomputing, NLPCC
- Volunteer: ACL 2024
Miscellaneous
I love sports, especially basketball, tennis, bouldering, and skiing. I’m a big fan of the Golden State Warriors, Stephen Curry, and Roger Federer. I enjoy reading books, especially science fiction and psychology. Some of my favorites include Narcissus and Goldmund and The Three-Body Problem. I also love watching movies. Some of my favorites are Fight Club by David Fincher and A Confucian Confusion by Edward Yang.