arXiv：cs.AI· Riccardo Barbano, Vincent Pauline, Runchang Li, George Webber, Alexander Denker, \v{Z}eljko Kereta, Stefan Bauer, Francisco Vargas, Esmeralda S. Whitammer·· 6 小时前AI 评分32
CMDS：用随机最优控制协调多个冻结扩散模型的多智能体扩散引导
One for All, All for One: Coordinated Multi-Agent Diffusion Steering via Stochastic Optimal Control
AI 导读
Coordinated Multi-Agent Diffusion Steering（CMDS）将多个冻结的预训练扩散模型视为可复用的生成基元，通过学习的控制信号协调其反向过程，把协调问题建模为随机最优控制。该方法能恢复已知目标分布、用同一控制满足不同空间约束，并从退化混合中恢复单个来源。在多智能体迷宫导航、关节机器人规划与文本条件人体动作任务上，CMDS 将冻结模型转化为协调的多智能体生成器。
正文
Abstract:Deep generative models often produce structured outputs composed of interacting components. Modelling these outputs with a single model requires learning both the component distributions and their interactions. We pursue a modular alternative: reuse independently trained component generators and learn only how to coordinate them to produce coherent structured outputs. Our framework, Coordinated Multi-Agent Diffusion Steering (CMDS), treats frozen pretrained diffusion models as reusable generative primitives and coordinates their reverse processes through a learned control. We formulate coordination as a stochastic optimal control problem, balancing an assembly-level reward that specifies the desired properties of the combined output against deviations from the pretrained dynamics. The learned control amortises this optimisation, allowing reuse across new task instances. Experiments show that CMDS can recover a known target distribution, satisfy different spatial constraints with the same trained control, and recover individual sources from degraded mixtures. Across multi-agent maze navigation, articulated robot planning, and text-conditioned human motion, CMDS turns frozen models into coordinated multi-agent generators.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.08595 [cs.RO] |
| (or arXiv:2610.08595v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08595 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Riccardo Barbano [view email]
[v1]
Tue, 6 Oct 2026 16:02:49 UTC (13,326 KB)
来源：arXiv：cs.AI · arxiv.org