Incoming graduate student at the Institute of Trustworthy Embodied Artificial Intelligence (TEAI), Fudan University. Currently a computer science undergraduate at China University of Mining and Technology. From December 2025 to August 2026 I was a research intern at Tsinghua Shenzhen International Graduate School.
Homepage · Blog · Email · GitHub
My research asks how vision-language models can reason about the physical world and act over long horizons, when visual evidence is grounded in 3D geometry and carried across an episode as persistent memory.
My current focus is agentic robot learning and embodied reasoning — vision-language models that reason about the physical world and drive robots through long-horizon manipulation tasks. Longer term, I want to work on RSI (recursive self-improvement) in Embodied AI.
Alongside the algorithmic work, I have hands-on experience with physical platforms: teleoperation data collection and π0 fine-tuning on an SO-101 arm, and locomotion policy training for Unitree G1 / Go2 in MuJoCo.
- 2026.10 — Awarded the National Scholarship (2025–2026 academic year).
- 2026.09 — Admitted to the Institute of Trustworthy Embodied Artificial Intelligence (TEAI), Fudan University, as an incoming graduate student.
- 2026.07 — RoboStream accepted to ECCV 2026.
- 2026.07 — CALO accepted to IEEE Transactions on Cloud Computing.
- 2026.03 — Conscious Gaze accepted to ICME 2026.
RoboStream: Weaving Spatio-Temporal Reasoning with Memory in Vision-Language Models for Robotics
Yuzhi Huang*, Jie Wu*, Weijue Bu*, Ziyi Xiong, Gaoyang Jiang, Ye Li, Kangye Ji, Shuzhao Xie, Yue Huang, Chenglei Wu, Jingyan Jiang, Zhi Wang
ECCV 2026 · Project Page · Code · arXiv
Conscious Gaze: Adaptive Attention Mechanisms for Hallucination Mitigation in Vision-Language Models
Weijue Bu, Guan Yuan, Guixian Zhang
ICME 2026 · arXiv
CALO: Code-Aware and Load-Aware Resource Configuration for Serverless Functions via Deep Reinforcement Learning
Donghong Xu, Weijue Bu, Zhouliang Ye, Shilong Wu
IEEE Transactions on Cloud Computing, 2026 · Code
ChainVLA: Chaining Vision-Language-Action Queries through a Unified Execution State for Long-Horizon Manipulation
Under review · Project Page
PiN-Mod: Identity-Consistent Personalized Image Generation
Under review
* Equal contribution.
Research Intern, Tsinghua Shenzhen International Graduate School
2025.12 – 2026.08
Long-horizon robotic manipulation and vision-language-model-based planning. Built the RLBench / SIMPLER evaluation pipeline and designed long- and short-horizon task suites; co-developed RoboStream (ECCV 2026) and worked on VLA execution-state memory.
- Fudan University — Institute of Trustworthy Embodied Artificial Intelligence (TEAI), incoming graduate student
- China University of Mining and Technology — undergraduate in computer science, 2023 – present
- National Scholarship (2025–2026 academic year), 2026
- Second Prize, 15th "China Software Cup" Collegiate Software Design Contest, National Finals, 2026
- National First Prize, "Challenge Cup" Competition, 2025
- 15th Place, Tianchi Alibaba Mobile Recommendation Algorithm Challenge, 2025
- Academic Potential 1st Prize, SJTU Summer School, 2025
- Second Prize, RAICOM Jiangsu Division Programming Skills Competition, 2025
- Bronze Medal, ICPC China Invitational, 2025
- Second Prize (Provincial), Lanqiao Cup C/C++ Programming, 2025
- Second Prize (Provincial), Jiangsu Higher Mathematics Competition, 2024
- Third Prize (National), GPLT Group Programming Ladder Tournament, 2024
PyTorch · JAX · MuJoCo · ROS · RLBench · LIBERO · SIMPLER · Docker · Linux