VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model

TL;DR

VLA-ATTC enhances vision-language-action models with adaptive test-time computation and a novel relative action critic, significantly reducing failure rates in complex embodied manipulation tasks.

Contribution

It introduces an uncertainty-based cognitive clutch and a relative action critic for improved decision-making in VLA models, with automated data curation and efficiency strategies.

Findings

Reduced failure rate of SOTA model PI0.5 by over 50% on LIBERO-LONG.

Introduced a relative action critic that simplifies learning objectives.

Developed an automated data pipeline for preference pair generation.

Abstract

Vision-Language-Action (VLA) models have demonstrated remarkable capabilities and generalization in embodied manipulation. However, their decision-making relies on a fast, instinctive process that lacks deliberation. This strategy often leads to suboptimal or catastrophic actions when facing complex or ambiguous scenarios that require greater consideration. In this paper, we introduce \textbf{VLA-ATTC}, a framework that endows VLA models with adaptive test-time compute (TTC). VLA-ATTC employs an uncertainty-based ``cognitive clutch'' to dynamically transition from reflexive execution to a TTC deliberation phase when necessary. During TTC phase, a novel \textbf{Relative Action Critic} (RAC) model identifies the optimal action from generated candidates via pairwise comparisons. This relative mechanism replaces unstable absolute value estimation, significantly simplifying the learning objective. Furthermore, we introduce an efficient sampling strategy to amortize computational costs and an automated data pipeline that curates preference pairs without manual annotation. On the LIBERO-LONG benchmark, VLA-ATTC reduces the failure rate of the SOTA model PI0.5 by over 50\%. We will open-source all the code and weights.