About me
Hi! I am Ke Ji (冀轲), a PhD student at School of Data Science, The Chinese University of Hong Kong, Shenzhen, advised by Prof. Benyou Wang. Prior to CUHK(SZ), I earned my master’s from Southeast University, supervised by Prof. Peng Wang, and my bachelor’s from Zhengzhou University.
My research centers on self-improving and trustworthy LLM agents — building agents that can autonomously evolve their reasoning capabilities and remain reliable in complex real-world deployments. Concretely, I work on: (1) agentic reasoning and efficient inference-time adaptation; (2) LLM self-improvement without human supervision; and (3) trustworthy evaluation of agents in real-world tasks.
Find me on Google Scholar, and Github!
Email: keji@link.cuhk.edu.cn
📝 Selected Publications
* denotes co-first authors
-
The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models
Ke Ji*, Jiahao Xu*, Tian Liang*, Qiuzhi Liu*, Zhiwei He, Xingyu Chen, et al.
NeurIPS 2025 -
Unlocking LLMs’ Self-Improvement Capacity with Autonomous Learning for Knowledge Generalization
Ke Ji, Junying Chen, Anningzhe Gao, Wenya Xie, Xiang Wan, Benyou Wang
ACL 2025 Findings -
Do Phone-Use Agents Respect Your Privacy?
Zhengyang Tang*, Ke Ji*, Xidong Wang*, Zihan Ye*, et al.
arxiv preprint -
Hierarchical Verbalizer for Few-Shot Hierarchical Text Classification
Ke Ji, Yixin Lian, Jingsheng Gao, Baoyuan Wang
ACL 2023 -
Domain-Hierarchy Adaptation via Chain of Iterative Reasoning for Few-shot Hierarchical Text Classification
Ke Ji, Peng Wang, Wenjun Ke, Guozheng Li, Jiajun Liu, Jingsheng Gao, et al.
IJCAI 2024 -
Enhancing Persona Consistency for LLMs’ Role-Playing using Persona-Aware Contrastive Learning
Ke Ji, Yixin Lian, Linxu Li, Jingsheng Gao, Weiyuan Li, Bin Dai
ACL 2025 Findings -
WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
Junteng Liu, Yunji Li, Chi Zhang, Jingyang Li, Aili Chen, Ke Ji, et al.
MiniMax Agent Technical Report -
The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence
MiniMax Team
MiniMax Technical Report -
SmartRAG: Jointly Learn RAG-Related Tasks From the Environment Feedback
Jingsheng Gao, Linxu Li, Ke Ji, Weiyuan Li, Yixin Lian, Yuzhuo Fu, et al.
ICLR 2025 -
HuatuoGPT-II: One-Stage Training for Medical Adaption of LLMs
Junying Chen, Xidong Wang, Ke Ji, Anningzhe Gao, et al.
COLM 2024 -
Agentifying Patient Dynamics within LLMs through Interacting with Clinical World Model
Minghao Wu, Yuting Yan, Zhenyang Cai, Ke Ji, et al.
arxiv preprint -
QFFT: Question-Free Fine-Tuning for Adaptive Reasoning
Wanlong Liu, Junxiao Xu, Fei Yu, Yukang Lin, Ke Ji, Wenyu Chen, et al.
NeurIPS 2025 (Spotlight) -
HuatuoGPT-o1: Towards Medical Complex Reasoning with LLMs
Junying Chen, Zhenyang Cai, Ke Ji, Xidong Wang, Wanlong Liu, Rongsheng Wang, et al.
ACL 2025 Findings -
CoD: Towards an Interpretable Medical Agent using Chain of Diagnosis
Junying Chen*, Chi Gui*, Anningzhe Gao, Ke Ji, et al.
ACL 2025 Findings
📖 Educations
- 2024.09 - now, PhD, The Chinese University of Hong Kong, Shenzhen.
- 2021.09 - 2024.07, Master, Southeast University, Nanjing. (National Scholarship, Top 1%)
- 2017.09 - 2021.06, Bachelor, Zhengzhou University, Zhengzhou. (GPA Rank: Top 5%)
🎖 Honors and Awards
- AIMO2 Gold Medalist in 2025 (HUAWEI-CUHKSZ)
- Southeast University Outstanding Graduate in 2024
- National Scholarship in 2023
- Excellent Graduate Cadre and Pacemaker to Merit Student of Southeast University in 2023
- Zhengzhou University Outstanding Graduate in 2021
🎓 Teaching Experience
- 2025.02 - 2025.07, TA of Natural Language Processing, The Chinese University of Hong Kong, Shenzhen.
- 2024.09 - 2025.01, TA of Large Language Models, The Chinese University of Hong Kong, Shenzhen.
💻 Internships
- 2025.06 - 2025.12, Interleaved Agent, MiniMax.
- 2024.10 - 2025.05, Self-Improvement and Mathematical Reasoning, Tencent AI Lab.
- 2022.01 - 2023.01, Prompt-based Learning, Xiaoice.