Jianfeng Cai (蔡建峰)
I am pursuing a master's degree in Information and Communication Engineering at the University of Science and Technology of China (USTC), where I work under the supervision of Prof. Houqiang Li and Prof. Wengang Zhou at the Microsoft Key Laboratory of Multimedia Computing and Communication.
My research focuses on the post-training of large language models, with a particular interest in reinforcement learning and coding agents. I welcome conversations about related research and potential collaborations.
I expect to graduate in fall 2027 and am seeking full-time roles focused on LLM post-training and coding agents. Please feel free to get in touch about relevant opportunities.
Publications
Kimi K3: Open Frontier Intelligence
Bias Fitting to Mitigate Length Bias of Reward Model in RLHF
Disentangling Length Bias In Preference Learning Via Response-Conditioned Modeling
CodeContests-O: Powering LLMs via Feedback-Driven Iterative Test Case Generation
Multi-Level Aware Preference Learning: Enhancing RLHF for Complex Multi-Instruction Tasks
Internships
Moonshot AI (月之暗面)
Research Intern RL Team
Kimi Model — Coding Agent Track
Ant Group
Machine Learning Intern 百灵&数字科技
Ling-DT Agent Model
Institute of Artificial Intelligence, Hefei Comprehensive National Science Center
Machine Game Intelligence Group
Large Language Models
Education
Technical Skills
- Python
- C/C++
- PyTorch
- verl
- OpenRLHF
- DeepSpeed