Verify, Act, Distill: Turning Test-Time Compute into Test-Time Adaptation for Vision-Language-Action Models

Key points

• We propose Verify, Act, Distill, a framework that consolidates the gains of TTC into the VLA itself.1
• A fixed verifier scores candidates generated by a flow-matching VLA, the weighted aggregate is executed, and the observations and aggregated chunks from successful adaptation episodes are distilled directly into LoRA parameters with the standard flow-matching loss.1
• On the 40 LIBERO tasks with a SmolVLA 5k checkpoint, distilling from only 10 episodes per task raises the mean success rate of verifier-free single-candidate generation from 54.1% to 70.6%, retaining about 74% of the improvement obtained by aggregating eight candidates.1
• These results show that using verification as a learning signal, rather than only as an execution-time selector, both retains the improvements of TTC and re-amplifies the effect of subsequent TTC.1

Sources (1)

• [1]Verify, Act, Distill: Turning Test-Time Compute into Test-Time Adaptation for Vision-Language-Action Models
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 10, 12:57 PM

  Test-time compute (TTC), which scores multiple candidate actions at inference time, is a powerful way to improve the execution performance of vision-language-action (VLA) models, but existing methods keep the policy frozen, so the gain must be recomputed at every execution.

  We propose Verify, Act, Distill, a framework that consolidates the gains of TTC into the VLA itself.

Extractive summary: sentences quoted from the sources.