SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models

TL;DR

SCALE is an inference strategy for vision-language-action models that adaptively modulates perception and action based on self-uncertainty, improving robustness without extra training or multiple passes.

Contribution

It introduces SCALE, a novel single-pass, training-free method that jointly adjusts perception and action using self-uncertainty, addressing perceptual ambiguity in VLAs.

Findings

Outperforms existing TTS methods on benchmarks

Enhances robustness under perceptual ambiguity

Maintains single-pass efficiency

Abstract

Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic control, with test-time scaling (TTS) gaining attention to enhance robustness beyond training. However, existing TTS methods for VLAs require additional training, verifiers, and multiple forward passes, making them impractical for deployment. Moreover, they intervene only at action decoding while keeping visual representations fixed-insufficient under perceptual ambiguity, where reconsidering how to perceive is as important as deciding what to do. To address these limitations, we propose SCALE, a simple inference strategy that jointly modulates visual perception and action based on 'self-uncertainty', inspired by uncertainty-driven exploration in Active Inference theory-requiring no additional training, no verifier, and only a single forward pass. SCALE broadens exploration in both perception and action under high uncertainty, while focusing on exploitation when confident-enabling adaptive execution across varying conditions. Experiments on simulated and real-world benchmarks demonstrate that SCALE improves state-of-the-art VLAs and outperforms existing TTS methods while maintaining single-pass efficiency.