REACH: Explicit Recovery Behavior for Diffusion Policies
Zundong Ke, Junlin Chen, Jiayi Zhu, Kuanhao Xia, Boyi Zhao, Jiayuan Gu
2026年份
摘要
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.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
它引用的顶会 Paper17
- Denoising Diffusion Probabilistic ModelsJonathan Ho, Ajay Jain, Pieter AbbeelNeurIPS 2020 · 被引用 35,902 次
- Diffusion Models Beat GANs on Image SynthesisPrafulla Dhariwal, Alexander Quinn NicholNeurIPS 2021 · 被引用 13,211 次
- High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser 等CVPR 2022 · 被引用 13,123 次
- Denoising Diffusion Implicit ModelsJiaming Song, Chenlin Meng, Stefano ErmonICLR 2021 · 被引用 11,743 次
- GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion ModelsAlexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam 等ICML 2022 · 被引用 4,691 次
相关 Paper
- ManipDreamer3D: Synthesizing Plausible Robotic Manipulation Video with Occupancy-aware 3D TrajectoryYing Li, Xiaobao Wei, Xiaowei Chi, Yuming Li 等AAAI 2026
- Failure Prediction at Runtime for Generative Robot PoliciesRalf Römer, Adrian Kobras, Luca Worbis, Angela P. SchoelligNeurIPS 2025 · 被引用 47 次
- Self-Correcting Robot Manipulation via Gaussian-Splatted ForesightShaohui Pan, Yong Xu, Ruotao Xu, Zihan Zhou 等AAAI 2025 · 被引用 2 次
- Generate Subgoal Images Before Act: Unlocking the Chain-of-Thought Reasoning in Diffusion Model for Robot Manipulation with Multimodal PromptsFei Ni, Jianye Hao, Shiguang Wu, Longxin Kou 等CVPR 2024 · 被引用 4 次
- FLARE: A Failure-Aware Framework for Autonomous Correction and Recovery in Visual-Language Robotic ManipulationGanlong Zhao, Zijia Tang, Xingping Chen, Zhanghui Kuang 等CVPR 2026 · 被引用 10 次
