Wen-Han Hsieh (謝文瀚)

santa_cruz.png

Hi! I’m Wen-Han, an aspiring researcher at Berkeley Artificial Intelligence Research Lab (BAIR), where I’m advised by Trevor Darrell and Koushil Sreenath.

Previously, I obtained my B.S. in MechE at National Tsing Hua University, under the supervision of James Chang.

My research interests center around:

  • Robotics & Embodied AI
  • Dexterous Hand Manipulation
  • Vision-Language-Action Models

Every decision and experience I pursue is driven by a long-term commitment to becoming a leading AI researcher. I will be applying to Ph.D. and Master’s programs this fall to further this mission.

Feel free to contact me through Email or Linkedin, I am happy to coffee chat : )

Selected Publications

(*: Equal contribution. Co-first Author.)

Do What? Teaching Vision-Language-Action Models to Reject the Impossible
Wen-Han Hsieh*, Elvis Hsieh*, Dantong Niu, Trevor Darrell, Roei Herzig, David M. Chan
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025
pdf code page
Publication image
LISAt: Language-Instructed Segmentation Assistant for Satellite Imagery
Jerome Quenum*, Wen-Han Hsieh*, Tsung-Han Wu, Ritwik Gupta, Trevor Darrell, David M. Chan
Advanced in Neural Information Processing Systems (NeurIPS) , 2025
pdf code page
Publication image
Optimizing TD3 for 7-DOF Robotic Arm Grasping: Overcoming Suboptimality with Exploration-Enhanced Contrastive Learning
Wen-Han Hsieh, Jen-Yuan Chang
International Conference on Knowledge Innovation and Invention (ICKII), 2024
Best Paper Award
pdf code
Publication image

Selected Projects

TransLingo
Wen-Han Hsieh, Jerry Chang, Vivian Dai, Tommy Chen, Christine Huang, Kodo Lin
Google Hardware Product Sprint, 2023
Description:
  • Transparent display-based AI-powered virtual translation system
  • MediaPipe-based dynamic hand gesture tracking, real-time, contactless screen control
code page
Project image
TrashAI
Wen-Han Hsieh, Cheng-Wei Kao, Yu-Sheng Ma, Yu-Cheng Liu, Yi-An Lian
Mei-Chu Hackathon (梅竹黑客松), 2023
Second Prize @ Taiwan’s largest hackathon (梅竹大獎第二名)
Description:
  • SW: AI-driven smart trash with real-time waste recognition system, classification across 5 waste categories using CNN.
  • HW: MIMXRT1060-EVKB microcontroller classifies wastes, controls 2 servo motors for dumping waste into the corresponding trash can.
code page