ManipLVM-R1 applies RLVR with IoU and trajectory-distance rewards to train a 3B VLM for affordance perception and trajectory prediction, claiming better performance and generalization than SFT on 50% of the data.
Hershey John R, Olsen Peder A
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models
ManipLVM-R1 applies RLVR with IoU and trajectory-distance rewards to train a 3B VLM for affordance perception and trajectory prediction, claiming better performance and generalization than SFT on 50% of the data.