论文/arXiv 2610.14885·18 天前

Verify, Act, Distill：将测试时计算转化为视觉-语言-动作模型的测试时适应

该文针对VLA测试时计算收益需反复重算的问题，提出Verify, Act, Distill，将验证器聚合的成功执行监督蒸馏进LoRA，并在LIBERO任务上验证。

AI 整理

• 具身模型
• 端侧部署
• 评测与安全

原文链接：Verify, Act, Distill: Turning Test-Time Compute into Test-Time Adaptation for Vision-Language-Action Models（arXiv 2610.14885 · 发布日期：2026-10-10）

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

该文针对VLA测试时计算收益需反复重算的问题，提出Verify, Act, Distill，将验证器聚合的成功执行监督蒸馏进LoRA，并在LIBERO任务上验证。

核心要点

• 现有TTC在推理时评分多个候选动作，但策略冻结，每次执行都需重新计算收益。
• 方法用固定验证器评分flow-matching VLA候选，执行加权聚合，并将成功片段蒸馏到LoRA参数。
• 在40个LIBERO任务上，每任务10个episode蒸馏使单候选成功率由54.1%升至70.6%。
• 蒸馏后再用同样聚合可升至82.3%，单候选策略也在含分布偏移的LIBERO-plus上改进。

验证范围与局限：验证范围包括40个LIBERO任务、SmolVLA 5k检查点、每任务10个episode蒸馏及LIBERO-plus分布偏移设置。

英文摘要 (Abstract)

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. 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. This reuses supervision that TTC has already produced during execution, without additional reinforcement learning or rollouts dedicated to teacher generation. 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. Applying the same aggregation to the distilled policy further raises the success rate from 76.4% to 82.3%, with a particularly large gain on LIBERO-Long (44.0% to 57.8%), and the single-candidate policy also improves on LIBERO-plus, which contains distribution shifts. 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.

作者：Kaneyoshi Hiratsuka, Satoshi Yamamori, Jun Morimoto

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.14885：Verify, Act, Distill: Turning Test-Time Compute into Test-Time Adaptation for Vision-Language-Action Models访问 2026-10-21