ICML 2026 2026
About
Hi, I'm Bikang — a computer science Ph.D. student at ShanghaiTech University, supervised by Prof. Ye Shi (石野).
My research centers on Embodied AI — particularly world action models and reinforcement post-training for building agents that perceive, plan, and act in complex environments.
I'm broadly interested in policy learning, vision-language foundation models, and optimization. If any of this resonates, I'd love to hear from you — drop me an email to discuss or explore collaboration.
News
- 2026.04 Our paper on sample-efficient diffusion-based RL with critic guidance accepted to ICML 2026. ICML 2026
- 2025.08 Pivoted research focus to Embodied AI — Vision-Language-Action (VLA) models and reinforcement post-training.
- 2025.03 Our paper NLPrompt (noise-label prompt learning for vision-language models) accepted to CVPR 2025. CVPR 2025Highlight
- 2024.09 Our paper on federated learning from vision-language foundation models accepted to NeurIPS 2024. NeurIPS 2024
- 2023.06 Awarded "Shanghai Outstanding Graduate".
Selected Publications
CVPR 2025 2025
ICCV 2025 2025
NeurIPS 2024 2024
SGC 2024 2024
AAAI 2023 2023
Certifiable Out-of-Distribution Generalization
Honors & Services
- Honors & Awards Shanghai Outstanding Graduate (2023) · Outstanding Student Leader (2026) · Outstanding Student (2025) · SIST Excellent Teaching Assistant (2025) · MCM Outstanding Winner (2022) · Perfect GPA 4.0 (Fall 2021)
- Conferences & Journals IJCV · ICCV 2025 · CVPR 2025 · NeurIPS 2024 & 2025 (Top Reviewer)
Education
- ShanghaiTech University · China Ph.D. in Computer Science and Technology
- ShanghaiTech University · China B.Eng. in Computer Science and Technology
Teaching
Teaching Assistant at ShanghaiTech University (SIST).
- SI251 Convex Optimization (凸优化)
- SI152 Numerical Optimization (数值最优化)
- EE140 Communication Principles (通信原理)
- MATH1122 Linear Algebra II (线性代数II)