ROAD-VLA: 自己蒸留による視覚言語行動モデルのロバストなオンライン適応
ROAD-VLA: Robust Online Adaptation via Self-Distillation for Vision-Language-Action Models
ロボット操作のための視覚言語行動モデル(VLA)のオンライン適応を強化するため、行動空間で近位教師を構築する自己蒸留フレームワークを提案。スパースな報酬を密なトークンレベルの信号に変換し、PPOを上回る性能を示した。
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: Kejing Wang, Toan Nguyen, Minh Hoang Nguyen, Simon Khan, Flora D. Salim
分類: cs.LG, cs.RO
原文アブストラクト
Effective online adaptation of vision-language-action (VLA) models remains challenging, as sparse rewards provide weak supervision for high-dimensional autoregressive action policies. Although self-distillation can in principle provide denser training signals, we find that text-based privileged teachers conditioned on demonstrations, retrieved experiences, or high-level plans are ineffective for VLA adaptation, exposing a modality gap between symbolic guidance and low-level robot actions. We propose ROAD-VLA, an advantage-guided self-distillation framework that constructs a proximal teacher directly in action space by perturbing action-token logits with calibrated advantage estimates. This converts sparse rewards into dense token-level supervision while keeping the teacher close to the current policy. We further derive a policy-improvement lower bound under calibrated advantages and accurate teacher matching. Across seven robotic manipulation environments with in-distribution and out-of-distribution shifts, ROADVLA outperforms PPO in nearly all settings, demonstrating robust online VLA adaptation.
