robotics
The Robot Denoises Whether It Is Done
Echo folds stop and failure signals into each action chunk, learning completion and recovery cues without frame-by-frame progress labels.
Summary
Echo folds stop and failure signals into each action chunk, learning completion and recovery cues without frame-by-frame progress labels.
A robot policy needs to know not only what to do next but whether the task has succeeded or failed. Echo adds stop and failure values as two dimensions of every vision-language-action chunk and denoises them alongside the motor commands. Successful, failed and recovery trajectories provide complementary supervision without frame-level progress labels. The status dimensions are denoised faster, allowing later action steps to condition on cleaner estimates while controlled interactions keep status and action learning coupled. The authors report accurate status prediction and consistent policy improvements in real-world manipulation experiments. The evidence belongs to the tested tasks, data construction and model design; it does not show that every failure will be detected.
Why it matters
Echo folds stop and failure signals into each action chunk, learning completion and recovery cues without frame-by-frame progress labels.
Limits and context
- A robot policy needs to know not only what to do next but whether the task has succeeded or failed.
- The evidence belongs to the tested tasks, data construction and model design; it does not show that every failure will be detected.
Key claims
Echo folds stop and failure signals into each action chunk, learning completion and recovery cues without frame-by-frame progress labels.
Qualification: A robot policy needs to know not only what to do next but whether the task has succeeded or failed.
Evidence: source-2026-10-25-003
Sources
- arXiv preprint 2610.31569, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.