Lune

CVPR2026Top-tier venue

REACH: Explicit Recovery Behavior for Diffusion Policies

Zundong Ke, Junlin Chen, Jiayi Zhu, Kuanhao Xia, Boyi Zhao, Jiayuan Gu

2026Year

Abstract

Figure 1. REACH is an error-aware framework with prompt-guided diffusion policy for Robot manipulation. A multi-camera setup observes the task execution along the timeline, while an error-detector monitors an anomaly score (bottom). When the score indicates an error, the robot rolls back to a previous safe action and the diffusion model regenerates a corrected prompt, allowing the robotic arm to continue the task and eventually succeed.

Ask about this paper

Your agent reads all of it.

Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.

Questions to start from

Your agent calls

Luneget_paper_fulltext

Ask in Lune

Free to start. No credit card required.

lune papers fulltext c3b02548-3cb7-4f21-bb48-6ea6b4d736ec

Builds on17

Related papers

Dusk over the sea between two cliffs drawn in fine vertical lines