在视觉-语言-动作对齐中，验证比策略学习更具效率
Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment
AI总结：
本文提出CoVer-VLA验证方法，通过测试时验证在视觉-语言-动作对齐中实现22%的在分布和13%的分布外收益，同时在现实世界实验中进一步提升45%。
摘要：
通用目的机器人的长期愿景依赖于其理解和执行自然语言指令的能力。视觉-语言-动作（VLA）模型在实现这一目标方面取得了显著进展，但生成的动作仍可能与给定的指令不一致。在本文中，我们研究了测试时验证作为缩小
英文摘要：
The long-standing vision of general-purpose robots hinges on their ability to understand and act upon natural language instructions. Vision-Language-Action (VLA) models have made remarkable progress toward this goal, yet their generated actions can still misalign with the given instructions. In this paper, we investigate test-time verification as a means to shrink the "intention-action gap." We first characterize the test-time scaling laws for embodied instruction following and demonstrate that jointly scaling the number of rephrased instructions and generated actions greatly increases test-time sample diversity, often recovering correct actions more efficiently than scaling each dimension independently. To capitalize on these scaling laws, we present CoVer, a contrastive verifier for vision-language-action alignment, and show that our architecture scales gracefully with additional computational resources and data. We then introduce CoVer-VLA, a hierarchical test-time verification pipeline using the trained verifier. At deployment, our framework precomputes a diverse set of rephrased instructions from a Vision-Language-Model (VLM), repeatedly generates action candidates for each instruction, and then uses the verifier to select the optimal high-level prompt and low-level action chunks. Compared to scaling policy pre-training on the same data, our verification approach yields 22% gains in-distribution and 13% out-of-distribution on the SIMPLER benchmark, with a further 45% improvement in real-world experiments. On the PolaRiS benchmark, CoVer-VLA achieves 14% gains in task progress and 9% in success rate.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
