A zero-shot pipeline using GPT-4o to label object subparts and a quality diversity grasp archive to select task-conditioned grasps reports 73.6% weighted IoU against human-annotated grasp regions and 88% human preference in a small user study.
A review of robot learning for manipulation: Challenges, representations, and algorithms,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Task-Aware Robotic Grasping by evaluating Quality Diversity Solutions through Foundation Models
A zero-shot pipeline using GPT-4o to label object subparts and a quality diversity grasp archive to select task-conditioned grasps reports 73.6% weighted IoU against human-annotated grasp regions and 88% human preference in a small user study.