Junwei Yu

Hi! I'm a member of technical staff at Voio Inc, where I train foundation models for radiology, and a researcher affiliated with BAIR. I'm fortunate to collaborate with Prof. XuDong Wang, Prof. Trevor Darrell, and Prof. Adam Yala. I received Bachelor's degree in Computer Science and Mathematics from UC Berkeley.

I'm interested in building multimodal intelligence that can help humans in the real world, from healthcare to everyday tasks.

Email  /  Scholar  /  Github  /  Linkedin

profile photo

Selected Publications

UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity
Junwei Yu, Trevor Darrell, XuDong Wang
TMLR 2026
Paper | Project page | Code
Stateful Visual Encoders for Vision-Language Models
Zirui Wang, Junwei Yu, Adam Yala, David M. Chan, Joseph E. Gonzalez, Trevor Darrell
EMNLP 2026
MEDFMB Workshop at ECCV 2026 (Oral)
X-Reason Workshop at ECCV 2026 (Oral)
Paper | Project page | Code
Fast John Ellipsoid Computation with Differential Privacy Optimization
(Alphabetical Order)
Xiaoyu Li*, Yingyu Liang*, Zhenmei Shi*, Zhao Song*, Junwei Yu*
CPAL 2025 (Oral)
Paper

Experience

Voio: Member of Technical Staff
Dec 2025 - Present: Research Team
Amazon: Software Engineering Intern
Summer 2025: Member of Global Payment Tech
DATA C140: Probability for Data Science
Fall 2024: Tutor
Supervised by Prof. Ani Adhikari