UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action Models

TL;DR

UAOR is a training-free, plug-and-play module that improves vision-language-action models by reinjecting observation information based on uncertainty, enhancing action confidence without extra data or training.

Contribution

Proposes UAOR, a novel uncertainty-aware observation reinjection method that enhances VLA models without additional data, training, or modules.

Findings

Consistently improves diverse VLA models in simulation and real-world tasks

Eliminates need for extra observation cues or modules

Operates with minimal computational overhead

Abstract

Vision-Language-Action (VLA) models leverage pretrained Vision-Language Models (VLMs) as backbones to map images and instructions to actions, demonstrating remarkable potential for generalizable robotic manipulation. To enhance performance, existing methods often incorporate extra observation cues (e.g., depth maps, point clouds) or auxiliary modules (e.g., object detectors, encoders) to enable more precise and reliable task execution, yet these typically require costly data collection and additional training. Inspired by the finding that Feed-Forward Network (FFN) in language models can act as "key-value memory", we propose Uncertainty-aware Observation Reinjection (UAOR), an effective, training-free and plug-and-play module for VLA models. Specifically, when the current language model layer exhibits high uncertainty, measured by Action Entropy, it reinjects key observation information into the next layer's Feed-Forward Network (FFN) through attention retrieval. This mechanism helps VLAs better attend to observations during inference, enabling more confident and faithful action generation. Comprehensive experiments show that our method consistently improves diverse VLA models across simulation and real-world tasks with minimal overhead. Notably, UAOR eliminates the need for additional observation cues or modules, making it a versatile and practical plug-in for existing VLA pipelines. The project page is at https://uaor.jiabingyang.cn.