# Gaussian Distillation of Flow Policies

**Authors:** Youssef Saied, Ziyad Sheebaelhamd, Claire Vernade, François Fleuret  
**Lab:** University of Geneva · **Published:** 29 Oct 2026  
**Links:** [arXiv](https://arxiv.org/abs/2610.40271) · [PDF](https://arxiv.org/pdf/2610.40271) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2610.40271.html)

## Key numbers

- **0.02%**: …a flow policy's final layer with a Gaussian readout trained on just 0.02% of eligible states

## Abstract

Flow policies are among the strongest robot controllers and are well suited to modelling multimodal action distributions. Our analysis shows that the continuous-action distributions of the studied flow policies are well approximated by Gaussians. We propose the Gaussian Action Readout (GAR), which replaces a flow policy's final layer with a Gaussian readout trained on just 0.02% of eligible states, while keeping the rest of the network frozen. It samples with one action-network pass, gives closed-form likelihoods, and achieves near-teacher task success at up to twice the inference speed.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.