---
title: "Efficient Diffusion Policy Fine Tuning with Latent Noise Representation Bridging"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.07275v1"
permalink: "https://feed.astralor.com/items/itm_e175a2852accd153efa04ab3"
published: "2026-11-05T00:49:53.000Z"
score: 58
channel: "tech"
tags: ["RAG","Efficient Diffusion Policy","Fine Tuning","Latent Noise Representation","Bridging Pre-trained","Pre-trained","Prior","Latent Noise Linear"]
---

# Efficient Diffusion Policy Fine Tuning with Latent Noise Representation Bridging

## At a glance

Pre-trained diffusion and flow policies have emerged as powerful backbones for robotics, yet adapting them to novel data via reinforcement learning (RL) remains computationally expensive, sample inefficient and lacks th…

## Key points

- Pre-trained diffusion and flow policies have emerged as powerful backbones for robotics, yet adapting them to novel data via reinforcement learning (RL) remains computationally expensive, sample inefficient and lacks th…
- Prior approaches that fine-tune the initial noise bypass costly backpropagation through time but struggle with the "noise aliasing" problem: there does not exista natural link between the noise space value and the actio…
- This paper resolves these challenges by introducing the Latent Noise Linear Markov Decision Process (LNL-MDP).

## Full text

Pre-trained diffusion and flow policies have emerged as powerful backbones for robotics, yet adapting them to novel data via reinforcement learning (RL) remains computationally expensive, sample inefficient and lacks theoretical guarantees. Prior approaches that fine-tune the initial noise bypass costly backpropagation through time but struggle with the "noise aliasing" problem: there does not exista natural link between the noise space value and the action space value. This paper resolves these challenges by introducing the Latent Noise Linear Markov Decision Process (LNL-MDP). We derive the Bridge Equation for LNL-MDPs which eliminates the noise aliasing problem by aligning values with a shared representation between the noise and the action space. Furthermore, our LNL-MDP fine tuning framework falls under the FineTuneRL theory with near-optimal guarantee. Leveraging these insights, we propose Latent Representation Bridging for Diffusion (LARBRID), an algorithm that efficiently learns the representation to optimize the initial distribution that steers the pre-trained model to act optimally in the environment. It yields up to 2-3$\times$ performance increases on challenging control problems across a diverse suite of robotic locomotion and manipulation tasks. Code can be found at: github.com/Maincode-Research/LARBRID.git.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_e175a2852accd153efa04ab3