AI RESEARCH
Knowledge-Guided Manipulation Using Multi-Task Reinforcement Learning
arXiv CS.LG
•
ArXi:2603.24083v1 Announce Type: cross Experiments on a suite of manipulation tasks with occlusions, distractors, and layout shifts nstrate consistent gains over strong baselines: the knowledge-conditioned agent achieves higher success rates, improved sample efficiency, and stronger generalization to novel objects and unseen scene configurations.