Improving Robustness of Vision-Language-Action Models by Restoring Corrupted Visual Inputs

TL;DR

This paper identifies the vulnerability of vision-language-action models to visual corruptions and introduces CRT, a transformer-based restoration method, to improve their robustness without retraining the entire model.

Contribution

The paper proposes CRT, a plug-and-play vision transformer that restores corrupted visual inputs, significantly enhancing VLA models' robustness against sensor artifacts.

Findings

CRT restores visual inputs effectively under severe corruption.

VLA models maintain near-baseline success rates with CRT.

CRT does not require fine-tuning of the original models.

Abstract

Vision-Language-Action (VLA) models have emerged as a dominant paradigm for generalist robotic manipulation, unifying perception and control within a single end-to-end architecture. However, despite their success in controlled environments, reliable real-world deployment is severely hindered by their fragility to visual disturbances. While existing literature extensively addresses physical occlusions caused by scene geometry, a critical mode remains largely unexplored: image corruptions. These sensor-level artifacts, ranging from electronic noise and dead pixels to lens contaminants, directly compromise the integrity of the visual signal prior to interpretation. In this work, we quantify this vulnerability, demonstrating that state-of-the-art VLAs such as $\pi_{0.5}$ and SmolVLA, suffer catastrophic performance degradation, dropping from 90\% success rates to as low as 2\%, under common signal artifacts. To mitigate this, we introduce the Corruption Restoration Transformer (CRT), a plug-and-play and model-agnostic vision transformer designed to immunize VLA models against sensor disturbances. Leveraging an adversarial training objective, CRT restores clean observations from corrupted inputs without requiring computationally expensive fine-tuning of the underlying model. Extensive experiments across the LIBERO and Meta-World benchmarks demonstrate that CRT effectively recovers lost performance, enabling VLAs to maintain near-baseline success rates, even under severe visual corruption.