ReViP: Mitigating False Completion in Vision-Language-Action Models with Vision-Proprioception Rebalance

TL;DR

ReViP introduces a vision-proprioception rebalancing framework and a benchmark suite to reduce false completions in vision-language-action models for robotic manipulation, significantly improving robustness and success rates.

Contribution

The paper presents the first False-Completion Benchmark Suite and a novel ReViP framework that uses progress-aware visual cues to mitigate modality imbalance in VLA models.

Findings

ReViP reduces false completion errors by 26% over baseline models.

The benchmark suite effectively evaluates robustness under controlled perturbations.

ReViP improves success rates on real-world robotic tasks.

Abstract

Vision-Language-Action (VLA) models have advanced robotic manipulation by combining vision, language, and proprioception to predict actions. However, previous methods fuse proprioceptive signals directly with vision-language features, resulting in state-dominant bias and \textbf{false completions} despite visible execution failures. We systematically analyze this failure mode, attributing it to modality imbalance, where policies overly rely on internal state progression and underuse visual evidence. To address this, we introduce the first \textbf{False-Completion Benchmark Suite}, featuring eight tasks with three controlled perturbations (\emph{Object Drop}, \emph{Distractor Swap}, \emph{Relayout}) to comprehensively evaluate false completion. Moreover, we propose \textbf{ReViP}, a novel VLA framework with \textbf{Vi}sion-\textbf{P}roprioception \textbf{Re}balance to enhance visual grounding and robustness under perturbations. The key insight is to introduce auxiliary \emph{progress-aware visual cues} to adaptively modulate the coupling between semantic perception and proprioceptive dynamics. Specifically, progress-aware visual cues are extracted by an external Task-Stage Observer, which performs task-relevant reasoning on real-time observations to drive task-stage feature-wise linear modulation, enhancing environmental awareness and mitigating state-driven errors. Extensive experiments show that ReViP effectively mitigates false completion and improves success rates over strong VLA baselines, achieving a \textbf{26\%} gain over $\pi_0$ model on our suite, with gains extending to LIBERO, RoboTwin 2.0, and real-world evaluations.