ARC: A Reasoning Recipe for Robot Foundation Models
ARC: A Reasoning Recipe for Robot Foundation Models
This research proposes ARC, a reasoning framework that improves zero-shot task performance of existing robot foundation models (RFMs) through structured reasoning traces, automatic labeling, and model adaptation. The method achieves state-of-the-art results on benchmark tasks like RoboLab-120 and MolmoSpaces, with up to 50 percentage point gains on RoboLab-Reasoning-50 and an 82.2 percentage point improvement in task success on real robots.
This paper introduces ARC, a novel reasoning framework that enhances zero-shot task performance of robot foundation models (RFMs) without requiring additional robot demonstrations or large-scale training. The approach combines reasoning traces, automatic labeling, and model adaptation strategies, achieving significant improvements on benchmark tasks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org