World modeling has emerged as an effective co-training objective for robot policies, giving rise to World Action Models (WAMs) that jointly predict actions and future states. However, most WAMs predict future states in the native representation space of pretrained visual backbones, resulting in high-dimensional targets...
Jie Chen, Ruofei Bai, Yu-Xin Cai et al.· 0 citations
Grasping is a fundamental robotic capability that bridges perception and physical task execution. This paper studies grasp pose recovery under a perception-to-execution mismatch, where a grasp generated from visual perception may become spatially stale if the object moves before execution, using only sparse tactile int...
Hao-Ran Wang, Yu-Teng Sun, Yuan-Jie Li et al.· 0 citations
This work proposes TIDAL (Temporally Interleaved Diffusion and Action Loop), a hierarchical framework that decouples semantic reasoning from high-frequency actuation in large-scale Vision-Language-Action models and introduces a temporally misaligned training strategy to handle the resulting latency shift.
Yuteng Sun, Hao-Ran Wang, Ruofei Bai et al.· arXiv.org· 5 citations
This work presents VistaVLA, a novel two-stage framework that constructs a geometry- and semantics-aware 3D cognitive representation from 3D Gaussian primitives and grounds it as compact context tokens for VLA policy learning, and introduces Merge-then-Query (MtQ), a token summarization mechanism.
