EmbodyAgent: 蒸留・SFT・RLによる身体性VLMエージェントの事後学習
EmbodyAgent: Post-Training Embodied VLM Agents via Distillation, SFT, and RL
ブラックボックス教師からのデータ合成、SFTによる蒸留、オンラインRLを組み合わせ、オープンウェイトVLMをロボット操作エージェントへと事後学習するフレームワークを提案。LIBERO Spatialで成功率が0/100から14/100へ向上。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Wei Dai, Jinkun Liu, Jiayi Chen, Han Liu, Wenbo Ding, Shaolun Huang
分類: cs.RO
原文アブストラクト
Vision-language models (VLMs) can act as closed-loop robot-use agents (RUAs) by writing and executing control code through tool interfaces. However, turning an open-weight VLM into a high-performing embodied agent requires a complete post-training stack: out-of-the-box open VLMs fail completely at interactive tool contracts, strong frontier teachers are accessible only as black-box endpoints without exposed weights or token probabilities, and supervised imitation alone plateaus under rollout distribution shift. We present EmbodyAgent, an end-to-end post-training framework with three tightly coupled pipelines: (1) a black-box data synthesis pipeline that queries a strong teacher under a forced think-then-code contract and filters trajectories by task success; (2) an SFT pipeline that distills the explanation-and-program targets into an open-weight 9B VLM with both vision and language encoders trainable; and (3) an online RL pipeline that optimizes student-generated episodes via group-relative advantage with a runtime-error-penalized outcome reward. All three pipelines share one primitive-free Python execution tool exposing only raw visual observation and seven-dimensional end-effector control, with no action decoder and no interface shift across stages. Under a strictly separated evaluation protocol on ten LIBERO Spatial tasks (disjoint state splits, greedy test-time decoding, training-only safeguards), held-out success on 100 test states rises from 0/100 (base 9B VLM) to 14/100 (after SFT) and 84/100 (after RL), while training rollout success climbs from 8.4% to 87.7% and median model calls per episode drop from 36 to 11. We further identify a self-reinforcing runtime-error mode induced by outcome-based RL on executable code, and establish a failure-category monitoring and sampling-side curation protocol that stabilizes training.
