Ruoqu Chen

Imagined failures in a world model can pinpoint exactly which robot skill components need new demonstrations, boosting success from 13% to 75% with minimal data.

Purely autoregressive VLAs no longer require awkward hybrid diffusion heads to achieve fine-grained robot control: structuring discrete action tokenization as causally annealed flow matching recovers continuous diffusion precision while preserving end-to-end token efficiency.

Weave is presented, a unified framework for learning whole-body dexterous humanoid-object interaction from captured human demonstrations that converts captured human-object interactions into executable robot-object references through contact-aware retargeting and approach-motion completion.

R2S-Eval reveals that automated evaluation can outperform manual success counting by capturing nuanced differences in robot behavior quality.