CT-VAM: 小脳-視床に着想を得た効率的な視覚運動制御のための視覚行動モデル
CT-VAM: A Cerebello-Thalamic-Inspired Vision-Action Model for Efficient Visuomotor Control
大規模言語モデルによる高次推論とローカルでの高速制御を分離するため、小脳-視床経路に着想を得た軽量な視覚行動モデルCT-VAMを提案。68Mパラメータで大規模VLAに匹敵する性能を達成し、実機でも高速制御を実現した。
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: Jiacheng Li, Yize Guo, Jiabin Guo, Qingchen Liu, Jiahu Qin
分類: cs.RO, cs.AI
原文アブストラクト
Vision-language-action models have shown strong promise for robot manipulation, yet raw language is primarily needed to specify task intent rather than to be repeatedly processed during high-frequency low-level execution. Motivated by this separation, we propose a cerebello-thalamic-inspired vision-action model (CT-VAM) for efficient task-conditioned visuomotor control. CT-VAM acts as a compact local execution policy that predicts action chunks from dualview visual observations, proprioception, and a lightweight task condition, potentially enabling a practical cloud-edge paradigm in which high-level semantic reasoning can be handled by large models while fast closed-loop control runs on local hardware. To fuse heterogeneous inputs effectively, CT-VAM introduces TARS (Thalamic Action Routing Stream), a stream-separated conditional attention decoder that independently routes action, visual and task streams, preventing dense sensory tokens from overwhelming compact task-relevant conditions. With only 68M parameters, CT-VAM achieves LIBERO success rates competitive with substantially larger VLA models, while reducing inference latency. Together with flow-consistent inpainting for asynchronous chunk execution, CT-VAM supports high-frequency control and demonstrates robust realworld deployment on resource-constrained robotic platforms.
