A reward-model-free offline RL method that uses VLM-generated pairwise preferences and contrastive preference learning to train manipulation policies.
Champion-level drone racing using deep reinforce- ment learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Policy Learning from Large Vision-Language Model Feedback without Reward Modeling
A reward-model-free offline RL method that uses VLM-generated pairwise preferences and contrastive preference learning to train manipulation policies.