粗から精密へ：視覚言語行動モデルのためのアクショントークン計画
Coarse-to-Control: Action-Token Planning for Vision-Language-Action Models
VLAモデルに、将来の軌道を要約する粗いアクショントークンの計画を導入し、その計画に基づいて実行用のアクショントークンを生成する方式を提案。長期的なタスクでの性能が向上した。
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: Jinhao Wu, Shiduo Zhang, Yicheng Liu, Xiaopeng Yu, Sixian Li, Siyin Wang, Hang Zhao, Jing Huo, Yang Gao, Jingjing Gong, Xipeng Qiu, Yu-Gang Jiang
分類: cs.RO
原文アブストラクト
Most vision-language-action (VLA) models map observations directly to actions without explicit intermediate planning, which limits performance on long-horizon tasks where early mistakes compound. We propose Coarse-to-Control, a plan-execute VLA that introduces planning natively in the action-token space. The key idea is to let the policy first predict a compact sequence of coarse action tokens that summarize the intended future trajectory, and then generate executable action tokens conditioned on this plan. Because both planning and execution share a unified discrete action vocabulary, the plan stays close to the control manifold and provides directly actionable guidance rather than an abstract hint that must be translated back to motor commands. Experiments on LIBERO, SimplerEnv-WidowX, and real-world manipulation tasks show that action-token planning consistently improves over direct action generation, with the largest gains on long-horizon multi-stage tasks.
