RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation

原文：RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation

作者：Xuhui Kang, Sung-Wook Lee, Yan Zhuang, Tianle Zhong, Yen-Ling Kuo

来源：arXiv cs.CV（计算机视觉）

正文

Computer Science > Robotics

arXiv:2610.13578v1 (cs)

[Submitted on 9 Oct 2026]

Title:RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation

Authors:Xuhui Kang, Sung-Wook Lee, Yan Zhuang, Tianle Zhong, Yen-Ling Kuo

View a PDF of the paper titled RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation, by Xuhui Kang and 4 other authors

View PDF

HTML (experimental)

Abstract:Pre-trained robot policies, such as Vision-Language-Action (VLA) models, provide a strong foundation for generalist robotic manipulation, but adapting them to new tasks with only a few demonstrations remains challenging. Directly fine-tuning a pre-trained policy often achieves strong in-distribution performance, yet it can distort the representations learned during pre-training and weaken the model’s generalization ability under out-of-distribution shifts. We propose Retrieval-Anchored Representation Alignment (RARA), a two-stage adaptation method, to address this issue. First, RARA retrieves representation features from pretraining with similar task intent and motion, and aligns the policy’s representations of the target demonstrations with the pre-trained representations of the retrieved demonstrations. Second, RARA fine-tunes the policy on the target demonstrations while limiting how much its representations of the retrieved demonstrations change during fine-tuning. By reducing the representation mismatch before action learning and preserving pre-trained knowledge during adaptation, RARA limits unnecessary representation drift during fine-tuning. Experiments with Diffusion Policy and GR00T on 13 tasks in RoboTwin, RoboCasa, and on a real robot show that RARA improves the average success of fine-tuning by 9.5 points in-distribution and 13.6 points out-of-distribution, and outperforms all compared baselines. Project page: this https URL

Comments:

21 pages, 11 figures, 7 tables

Subjects:

Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)

Cite as:

arXiv:2610.13578 [cs.RO]

(or

arXiv:2610.13578v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2610.13578

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

主题

机器人 · 计算机视觉 · 机器学习

由「前沿雷达」于 2026-10-09 采集。正文取自原文页面，已保留出处链接。标题与正文版权归原作者所有。