Vision-Language-Action (VLA) models map visual observations and natural-language instructions to robot actions; however, hierarchical and autoregressive paradigms often incur architectural overhead...
<p><strong>Tác giả:</strong> Yang Liu, Pengxiang Ding, Tengyue Jiang, Xudong Wang, Wenxuan Song, Minghui Lin, Han Zhao, Hongyin Zhang, Zifeng Zhuang, Wei Zhao, Siteng Huang, Jinkui Shi, Donglin Wang</p>
<p>Vision-Language-Action (VLA) models map visual observations and natural-language instructions to robot actions; however, hierarchical and autoregressive paradigms often incur architectural overhead, accumulate long-horizon errors, and require auxiliary modules to capture environment dynamics. To this end, we present MMaDA-VLA, a fully native, pretrained discrete diffusion VLA that unifies multi-modal understanding and generation. Specifically, MMaDA-VLA uses a shared discrete token space to j...</p>
<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3767308.3836069" target="_blank" rel="noopener">https://doi.org/10.1145/3767308.3836069</a></p>
<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3767308.3836069" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>
Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex
