Chuxue Cao

👋 Hi! I am a second-year PhD student at the Hong Kong University of Science and Technology, advised by Prof. Sirui Han and Prof. Yike Guo. My research focuses on AI safety & alignment and formal reasoning.👋 哈喽！我是香港科技大学的二年级博士生，师从韩斯睿教授和郭毅可教授。我的研究方向为大模型安全对齐（LLM Safety & Alignment）与形式化推理（Formal Reasoning）。

🔍 Research🔍 研究方向

I study how language models can reason more reliably and safely. My work connects natural language reasoning with formal logic verification, aiming to build agentic post-training frameworks and benchmarks that push LLMs toward stronger logical reasoning and safe alignment.我主要研究大模型如何进行更可靠、安全的推理。我的工作将自然语言推理与形式化逻辑验证相结合，致力于后训练框架和安全评测基准，推动大模型向更可靠的逻辑推理与安全对齐发展。

🌟 Selected Publications🌟 代表性论文

ICML 2026

Pushing the Boundaries of Natural Reasoning: Interleaved Bonus from Formal-Logic Verification in Language Models [Paper]

Chuxue Cao*, Jinluan Yang*, Haoran Li, Kunhao Pan, Zijian Zhao, Zhengyu Chen, Yuchen Tian, Lijun Wu, Conghui He, Sirui Han✉, Yike Guo✉

A formal logic verification-guided framework that interleaves symbolic feedback with natural language generation for stronger LLM reasoning.

Technical Report, Shanghai AI Lab

Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training [Paper]

Chuxue Cao, Honglin Lin, Zhanping Zhong, Xin Gao, Mengzhang Cai, Conghui He, Sirui Han✉, Lijun Wu✉

A study on data distillation and difficulty-aware data selection for financial LLMs.

EMNLP 2025

Towards Advanced Mathematical Reasoning for LLMs via First-Order Logic Theorem Proving [Paper]

Chuxue Cao*, Mengze Li*, Juntao Dai, Jinluan Yang, Zijian Zhao, Shengyu Zhang, Weijie Shi, Chengzhong Liu, Sirui Han✉, Yike Guo✉

Towards advanced mathematical reasoning through first-order logic theorem proving.

ACL Findings 2025

SafeLawBench: Towards Safe Alignment of Large Language Models [Paper]

Chuxue Cao*, Han Zhu*, Jiaming Ji, Qichao Sun, Zhenghao Zhu, Yinyu Wu, Juntao Dai, Yaodong Yang, Sirui Han✉, Yike Guo✉

A legal-perspective benchmark for safe alignment of large language models.

📝 Conference Papers📝 会议论文

• From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms [Paper]
  Jinghao Luo, Yuchen Tian, Chuxue Cao, Ziyang Luo, Hongzhan Lin, Kaixin Li, Chuyi Kong, Ruichao Yang, Jing Ma
  ACL Findings, 2026
• ScholarGPT: Fine-tuning Large Language Models for Discipline-Specific Academic Paper Writing [Paper]
  Chuxue Cao, Ziqing Yuan, Hailiang Chen
  Pacific-Asia Conference on Information Systems (PACIS), 2024
• Medical Scene Graphs and Reasoning [Paper]
  Chuxue Cao, Yiming He, Yuzhen Chen, Chunli Song, Hao Ling, Renchu Guan, Xiaoyue Feng
  IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2021

🚀 Preprints🚀 预印本

• LRAS: Advanced Legal Reasoning with Agentic Search [Paper]
  Yujin Zhou*, Chuxue Cao*, Jinluan Yang*, Lijun Wu, Conghui He, Sirui Han✉, Yike Guo✉
  arXiv, 2026
• Trustworthy Legal Reasoning: A Comprehensive Survey [Paper]
  Sirui Han, Zhizhuo Kou, Ruoxi Li, Yuyao Zhang, Yujin Zhou, Chuxue Cao, Han Zhu, Kunhao Pan, Haoran Li, Conghui He, Haitian Lu, Yike Guo
  Preprints, 2026
• MedInsightBench: Evaluating Medical Analytics Agents Through Multi-Step Insight Discovery in Multimodal Medical Data [Paper]
  Zhenghao Zhu, Chuxue Cao, Sirui Han, Yuanfeng Song, Xing Chen, Caleb Chen Cao, Yike Guo
  arXiv, 2025
• Measuring Hong Kong Massive Multi-Task Language Understanding [Paper]
  Chuxue Cao*, Zhenghao Zhu*, Junqi Zhu, Guoying Lu, Siyu Peng, Juntao Dai, Weijie Shi, Sirui Han✉, Yike Guo✉
  arXiv, 2025

🎓 Education🎓 教育背景

💼 Experience💼 工作经历

📬 Contact📬 联系我

I’m always open to discussing research-related topics. Feel free to reach me at 欢迎交流～📮ccaoai@connect.ust.hk.