# Train-Small Deploy-Large: Leveraging Diffusion-Based Multi-Robot Planning

arXiv: https://arxiv.org/abs/2604.06598 · HTML: https://mlsanity.com/paper/2604.06598/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models.
- Builds on 20 of its 26 references that are in the map.
- First posted on arXiv 8 Apr 2026.

## Placed in

- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)

## Builds on (most cited first, 20 of 20)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Video Diffusion Models](https://mlsanity.com/paper/2204.03458/) (2022)
- [Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms](https://mlsanity.com/paper/1911.10635/) (2019)
- [MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusion](https://mlsanity.com/paper/2306.03083/) (2023)
- [Graph Learning: A Survey](https://mlsanity.com/paper/2105.00696/) (2021)
- [A Survey of Learning in Multiagent Environments: Dealing with Non-Stationarity](https://mlsanity.com/paper/1707.09183/) (2017)
- [MADiff: Offline Multi-agent Learning with Diffusion Models](https://mlsanity.com/paper/2305.17330/) (2023)
- [VMAS: A Vectorized Multi-Agent Simulator for Collective Robot Learning](https://mlsanity.com/paper/2207.03530/) (2022)
- [Multi-Robot Motion Planning with Diffusion Models](https://mlsanity.com/paper/2410.03072/) (2024)
- [Simultaneous Multi-Robot Motion Planning with Projected Diffusion Models](https://mlsanity.com/paper/2502.03607/) (2025)
- [CoBL-Diffusion: Diffusion-Based Conditional Robot Planning in Dynamic Environments Using Control Barrier and Lyapunov Functions](https://mlsanity.com/paper/2406.05309/) (2024)
- [NavFormer: A Transformer Architecture for Robot Target-Driven Navigation in Unknown and Dynamic Environments](https://mlsanity.com/paper/2402.06838/) (2024)
- [Generative Multi-Agent Collaboration in Embodied AI: A Systematic Review](https://mlsanity.com/paper/2502.11518/) (2025)
- [SICNav-Diffusion: Safe and Interactive Crowd Navigation with Diffusion Trajectory Predictions](https://mlsanity.com/paper/2503.08858/) (2025)
- [Discrete-Guided Diffusion for Scalable and Safe Multi-Robot Motion Planning](https://mlsanity.com/paper/2508.20095/) (2025)
- [Multi-Agent Path Finding in Continuous Spaces with Projected Diffusion Models](https://mlsanity.com/paper/2412.17993/) (2024)
- [Review of Multi-Agent Algorithms for Collective Behavior: a Structural Taxonomy](https://mlsanity.com/paper/1803.05464/) (2018)
- [Beyond Local Views: Global State Inference with Diffusion Models for Cooperative Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2408.09501/) (2024)

## Abstract (arXiv)

Learning based multi-robot path planning methods struggle to scale or generalize to changes, particularly variations in the number of robots during deployment. Most existing methods are trained on a fixed number of robots and may tolerate a reduced number during testing, but typically fail when the number increases. Additionally, training such methods for a larger number of agents can be both time consuming and computationally expensive. However, analytical methods can struggle to scale computationally or handle dynamic changes in the environment. In this work, we propose to leverage a diffusion model based planner capable of handling dynamically varying number of agents. Our approach is trained on a limited number of agents and generalizes effectively to larger numbers of agents during deployment. Results show that integrating a single shared diffusion model based planner with dedicated inter-agent attention computation and temporal convolution enables a train small deploy-large paradigm with good accuracy. We validate our method across multiple scenarios and compare the performance with existing multi-agent reinforcement learning techniques and heuristic control based methods.