RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models

TL;DR

RoboAlign introduces a systematic training framework that enhances vision-language-action models by using zero-shot reasoning and reinforcement learning, significantly improving their performance on robotics benchmarks.

Contribution

It presents RoboAlign, a novel approach combining zero-shot reasoning and reinforcement learning to reliably improve embodied reasoning in multimodal-large-language models for robotics.

Findings

Achieves up to 106.6% performance improvement on real-world environments.

Uses less than 1% of data for RL-based alignment.

Significantly outperforms SFT baselines on major benchmarks.

Abstract

Improving embodied reasoning in multimodal-large-language models (MLLMs) is essential for building vision-language-action models (VLAs) on top of them to readily translate multimodal understanding into low-level actions. Accordingly, recent work has explored enhancing embodied reasoning in MLLMs through supervision of vision-question-answering type. However, these approaches have been reported to result in unstable VLA performance, often yielding only marginal or even negative gains. In this paper, we propose a more systematic MLLM training framework RoboAlign that reliably improves VLA performance. Our key idea is to sample action tokens via zero-shot natural language reasoning and refines this reasoning using reinforcement learning (RL) to improve action accuracy. As a result, RoboAlign bridges the modality gap between language and low-level actions in MLLMs, and facilitate knowledge transfer from MLLM to VLA. To validate the effectiveness of RoboAlign, we train VLAs by adding a diffusion-based action head on top of an MLLM backbone and evaluate them on major robotics benchmarks. Remarkably, by performing RL-based alignment after SFT using less than 1\% of the data, RoboAlign achieves performance improvements of 17.5\%, 18.9\%, and 106.6\% over SFT baselines on LIBERO, CALVIN, and real-world environments, respectively.