Residual Denoising Enables Sample-Efficient Multi-Agent Coordination on Demand

Published September 30, 2026 · Category: Robotics

Overview

arXiv:2609.32129v2 Announce Type: replace Abstract: Pretrained robot policies offer strong manipulation skills but are typically limited to single-agent settings, where a robot acts in isolation. In this work, we study how to adapt pretrained single-agent diffusion policies to multi-agent settings using minimal collaborative data, co-optimizing for two key objectives: high coordination performance and single-agent skill retention. To this end, we introduce ALTER, an adaptation method for coordination on demand: the adapted policy coordinates with other robots when deployed in a team while remaining capable of acting independently when operating alone. Execution is decentralized: each robot acts only on its own visual observations, without explicit inter-agent communication. Our method trains a coordination head that predicts a residual denoiser to transform single-agent behavior into coordinated multi-agent behavior when necessary while also preserving single-agent capabilities. To preserve single-agent capabilities, we augment a small number of collaborative demonstrations with self-distilled data generated by the base policy during training of the residual denoiser. In simulation, ALTER achieves higher coordination success over our baselines while retaining much higher source-skill retention. In our hardware experiments, we find similar trends where ALTER better co-optimizes for coordination success and single-agent skill retention than the baselines.

Source

Originally published at arxiv.org.

Related Articles

• Passive-Dynamic-Walking-Inspired Dynamics Guidance for Energy-Efficient Humanoid Locomotion
• SAKI: Skill Assembly and Kinematic Imitation from Human Videos for Long-Horizon Mobile Manipulation
• KPI: A Promptable Kernel for Physical Interaction on Humanoids