Bowen Fang
Ph.D. Student at Columbia University.
I am a Ph.D. student at Columbia University
, where I work with the Data Science Institute (Smart Cities Center). My research centers on understanding the behaviors behind post-training algorithms for large models, evaluating system resilience, and studying how agents can bootstrap meaningful search and competence in huge spaces with sparse signal and minimal useful prior structure.
I am a two-time Applied Scientist Intern at AWS AI , where my work centers on post-training for large models.
I hold an M.S. in Operations Research from Columbia University
and a Bachelor in Big Data Management and Applications, minor in Economics from Peking University .
This website serves as a central hub for my publications, projects, and professional activities.
news
| Apr 26, 2026 | I will return to Amazon Bedrock this summer as an Applied Scientist Intern on the Model Customization team. |
|---|---|
| May 27, 2025 | I will be starting my Applied Scientist Internship at AWS AI Labs this summer! |
| May 20, 2025 | I am honored to be a winner of the CS3 VALIDATE Accelerator program, which will provide continued funding for our work SINA. |
| Mar 06, 2025 | Our paper, Efficient Consistency Model Training for Policy Distillation in Reinforcement Learning, was accepted to the ICLR 2025 DeLTa Workshop as a poster presentation. |
| Aug 01, 2024 | I am excited to begin my Ph.D. studies at Columbia University. |
latest posts
| Oct 29, 2023 | Some Recent Advancement Around MuZero |
|---|---|
| Jun 20, 2023 | MPC with a Differentiable Forward Model: An Implementation with Jax |
| May 12, 2023 | Adding MuZero into RL Toolkits at Ease |
selected publications
- Decaying Budget Forcing: A Simple and Effective Reinforcement Learning Approach for Balancing Accuracy and Capacity in Mathematical ReasoningIn submission, 2026
-
Do Math Reasoning LLMs Help Predict the Impact of Public Transit Events?Under review at Transportation Research Part C (Special Issue: Foundation Models and Large Language Models in Urban Mobility), 2025arXiv preprint - Survey of Reasoning-Based Autonomous Driving in Mixed Traffic: An Offline–Online Two-Loop PerspectiveUnder review at IEEE Transactions on Intelligent Transportation Systems (Special Issue: AI-Empowered Automated Driving in Mixed Traffic: From Sensing, Perception, to Planning and Control), 2026
-
Efficient Consistency Model Training for Policy Distillation in Reinforcement LearningIn ICLR 2025 Workshop on Deep Generative Model in Machine Learning: Theory, Principle and Efficacy, 2025 -
Learn to Tour: Operator Design for Solution Feasibility Mapping in Pickup-and-delivery Traveling Salesman ProblemIn Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC), 2025 -
TraveLLM: Could you plan my new public transit route in face of a network disruption?In Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC), 2025