フローマッチング方策の強化学習ファインチューニングによる視覚言語行動モデルの改善
Reinforcement Fine-Tuning of Flow-Matching Policies for Vision-Language-Action Models
フローマッチングベースのVLAモデルに対し、重要度サンプリングを条件付きフローマッチング目的の変化で再定式化したFPOを提案し、オンライン強化学習による安定したファインチューニングを実現した。
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: Mingyang Lyu, Yinqian Sun, Erliang Lin, Huangrui Li, Ruolin Chen, Feifei Zhao, Yi Zeng
分類: cs.LG, cs.RO
原文アブストラクト
Vision-Language-Action (VLA) models such as OpenVLA, Octo, and $\pi_0$ have shown strong generalization by leveraging large-scale demonstrations, yet their performance is still fundamentally constrained by the quality and coverage of supervised data. Reinforcement learning (RL) provides a promising path for improving and fine-tuning VLAs through online interaction. However, conventional policy gradient methods are computationally infeasible in the context of flow-matching based models due to the intractability of the importance sampling process, which requires explicit computation of policy ratios. To overcome this limitation, we propose Flow Policy Optimization (FPO) algorithm, which reformulates importance sampling by leveraging per-sample changes in the conditional flow-matching objective. Furthermore, FPO achieves stable and scalable online reinforcement fine-tuning of the $\pi_0$ model by integrating structure-aware credit assignment to enhance gradient efficiency, clipped surrogate objectives to stabilize optimization, multi-step latent exploration to encourage diverse policy updates, and a Q-ensemble mechanism to provide robust value estimation. We evaluate FPO on the LIBERO benchmark and the ALOHA simulation task against supervised, preference-aligned, diffusion-based, autoregressive online RL, and $\pi_0$-FAST baselines, observing consistent improvements over the imitation prior and strong alternatives with stable learning under sparse rewards. In addition, ablation studies and analyses of the latent space dynamics further highlight the contributions of individual components within FPO, validating the effectiveness of the proposed computational modules and the stable convergence of the conditional flow-matching objective during online RL.
