Z-sight: 視覚言語行動と潜在世界モデリングの統合
$Z$-sight: Unifying Vision-Language-Action and Latent World Modeling
VLMの視覚トークン空間で直接未来を予測し、行動生成と統合したロボット基盤モデル。ピクセル再構成なしで高精度・高速な制御を実現。
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: Junjie Ye, Yiqi Zhao, Muchen Xu, Celina Shiyu Wang, Hao Dong, Ruohai Ge, Rong Xue, Benjamin J Kolligs, Aditya Dutt, Sasanka Nagavalli, Nassim A. Samad, Gaurav S. Sukhatme, Jyotirmoy V. Deshmukh, Yue Wang, Yuliang Guo
分類: cs.RO, cs.CV
原文アブストラクト
We introduce $Z$-sight, a unified robot foundation model that couples vision-language-action (VLA) and generative latent world modeling. Video-based world action models (WAMs) show the value of joint future-and-action generation for control, yet inherit the heavy cost of video generators. Recent latent-space policies avoid pixel generation, yet typically align intermediate features or rely on frozen target encoders, decoupling future prediction from the VLM's native feature space. $Z$-sight instead predicts the future directly in the visual token space in which its VLM perceives the present. Coupling a foresight expert and a continuous action expert with the VLM in a mixture-of-transformers, $Z$-sight jointly denoises future spatial tokens and actions at inference without pixel reconstruction or external encoders. Trained on human video and robot trajectories, $Z$-sight achieves 81.9% success on SIMPLE (+9.6 points over the strongest baseline) and 73.3% across six Unitree G1 tasks (+20.0 points) at 110 ms per chunk, while remaining competitive on tabletop benchmarks.
