Real-World Offline Reinforcement Learning from Vision Language Model Feedback
Sreyas Venkataraman∗, Yufei Wang∗, Ziyu Wang, Zackory Erickson†, David Held†
Submission to ICRA 2025
*Equal Contribution; †Equal Advising
Embedded Files
Summary Video
Summary Video
icra-video-final.mp4Real World Dressing Trials
Real World Dressing Trials
viper-gown-ours2249-2294.mp4Offline RL-VLM-F (ours)
viper-gown-dp3-1705-1729.mp4DP3 baseline
viper-green-ours-4525-4561.mp4Offline RL-VLM-F (ours)
viper-green-dp3-3675-3719.mp4DP3 baseline
viper-purple-ours-1143-1190.mp4Offline RL-VLM-F (ours)
viper-purple-dp3-355-385.mp4DP3 baseline
manikin-gown-ours-1370-1423.mp4Offline RL-VLM-F (ours)
manikin-gown-dp3-838-895.mp4DP3 baseline
manikin-purple-ours-1970-2027.mp4Offline RL-VLM-F (ours)
manikin-purple-dp3-2480-2544.mp4DP3 baseline
manikin-green-ours-1276-1321.mp4Offline RL-VLM-F (ours)
manikin-green-dp3-1139-1182.mp4DP3 baseline
Page updated
Report abuse