---
license: mit
language:
- en
tags:
- code
---
# Action Flow Matching (AFM)

Data and model checkpoints for the paper **"Action Flow Matching for Continual Robot Learning"** presented at Robotics: Science and Systems (RSS) 2025.

[![View on RSS](https://img.shields.io/badge/Read%20Paper-RSS-blue?logo=robotics&style=flat)](https://roboticsconference.org/program/papers/26/)
[![View on arXiv](https://img.shields.io/badge/Read%20Paper-arXiv%3A2504.18471-red?logo=arxiv&style=flat)](https://arxiv.org/abs/2504.18471)
[![View on GitHub](https://img.shields.io/badge/Code-GitHub-black?logo=github&style=flat)](https://github.com/AlejandroMllo/action_flow_matching/tree/main)

![Overview](https://github.com/AlejandroMllo/action_flow_matching/blob/main/assets/overview.jpg)

AFM enables efficient, non-episodic, online robot model refinement. AFM adjusts the planned actions seeking to reduce the gap between the intended and realized consequences of the action.