This paper addresses the inherent inference latency challenges associated with deploying multimodal large language models (MLLM) in quadruped vision-language-action (QUAR-VLA) tasks. Our investigation reveals that conventional parameter reduction techniques ultimately impair the performance of the language foundation model during the action instruction tuning phase, making them unsuitable for this purpose. We introduce a novel latency-free quadruped MLLM model, dubbed QUART-Online, designed to enhance inference efficiency without degrading the performance of the language foundation model. By incorporating Action Chunk Discretization (ACD), we compress the original action representation space, mapping continuous action values onto a smaller set of discrete representative vectors while preserving critical information. Subsequently, we fine-tune the MLLM to integrate vision, language, and compressed actions into a unified semantic space. Experimental results demonstrate that QUART-Online operates in tandem with the existing MLLM system, achieving real-time inference in sync with the underlying controller frequency, significantly boosting the success rate across various tasks by 65%. Our project page is https://quart-online.github.io.
@article{arxiv.2412.15576,
title = {QUART-Online: Latency-Free Large Multimodal Language Model for Quadruped Robot Learning},
author = {Xinyang Tong and Pengxiang Ding and Yiguo Fan and Donglin Wang and Wenjie Zhang and Can Cui and Mingyang Sun and Han Zhao and Hongyin Zhang and Yonghao Dang and Siteng Huang and Shangke Lyu},
journal= {arXiv preprint arXiv:2412.15576},
year = {2025}
}
Comments
Accepted to ICRA 2025; Github page: https://quart-online.github.io