用统一离散扩散模型实现视觉语言动作的端到端生成，提升长序列任务一致性。
MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation
- 采用共享离散标记空间，同步去噪目标图像与动作片段。
- 在LIBERO上达98.0%成功率，CALVIN上成功序列长度达4.78。
- 无需额外世界模型，适合复杂长程机器人任务部署。
视觉-语言-动作（VLA）模型将视觉观测和自然语言指令映射为机器人动作；然而，分层与自回归范式常带来架构开销，累积长时序误差，并需辅助模块捕捉环境动态。为此，我们提出MMaDA-VLA，一种完全原生的预训练离散扩散VLA模型，实现了多模态理解与生成的统一。具体而言，MMaDA-VLA使用共享的离散标记空间，联合去噪未来目标观测与动作块，使动作基于预测的视觉结果，无需辅助世界模型。该方法实现并行、无序的优化，提升长时序一致性。大量实验与全面分析表明，MMaDA-VLA在LIBERO上平均成功率达98.0%，在CALVIN上平均成功序列长度为4.78，且在真实场景中表现优异。项目页面见https://yliu-cs.github.io/MMaDA-VLA。
原文摘要 · Abstract (English)
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 jointly denoise a future goal observation and an action chunk, grounding actions in predicted visual outcomes without an auxiliary world model. In this way, parallel, order-free refinement improves long-horizon consistency. Extensive experiments and comprehensive analyses demonstrate that MMaDA-VLA achieves an average success rate of 98.0\% on LIBERO and an average successful sequence length of 4.78 on CALVIN, while performing strongly in real-world settings. The project page is available at https://yliu-cs.github.io/MMaDA-VLA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。