arXiv：cs.AI· Billy Yan, Isaac Tu, Kelly Lee, Blake Chang, Kenny Ma, Kanad Patel, Dhawal Kabra, Neer Patel, Kevin Wu, Lerrel Pinto, Irmak Guzey·· 20 小时前
Steertuning：用 5 段人类视频让机器人 5 分钟内学会新任务
Steertuning Policies from a Few Human Videos
AI 导读
Steertuning 提出一种机器人学习方法，仅需 5 段人类演示视频即可在新环境中学会新任务，从数据采集到策略部署全程不到 5 分钟。该方法先在 100 小时自我中心人类交互数据上预训练 400M 参数扩散策略，再对新任务用 2 分钟（2000 次训练更新）微调，并通过反向采样中选取观测最接近部署环境的演示来引导动作。
正文
Authors:Billy Yan, Isaac Tu, Kelly Lee, Blake Chang, Kenny Ma, Kanad Patel, Dhawal Kabra, Neer Patel, Kevin Wu, Lerrel Pinto, Irmak Guzey
Abstract:Current recipes for robot learning require massive amounts of robot demonstrations to obtain robust and adaptive policies. In this work, we introduce Steertuning, a system that enables robots to learn new tasks in new environments from 5 human demonstrations in under 5
Steertuning consists of two parts. (1) We pretrain a 400M-parameter diffusion policy on our egocentric human dataset, which consists of 100 hours of densely annotated random interactions. This pretraining makes Steertuning robust to background differences and enables spatial and object generalization when deployed on new tasks. (2) Given 5 demonstrations of a new task, we finetune our model for only 2 minutes (2000 training updates) and steer the policy's actions during reverse sampling using the demonstration whose observation is closest to that of the deployment environment. Together with a carefully designed data processing pipeline for embodiment transfer, this enables Steertuning to achieve a 71% success rate across 12 tasks in 12 new environments, ranging from lab settings to kitchens in new homes. The entire process, from data collection for a new task to robot policy deployment, takes less than 5 minutes. Our full dataset, model weights, codebase, and robot videos will be made available at our website: this https URL.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.18711 [cs.RO] |
| (or arXiv:2610.18711v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.18711 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Irmak Guzey [view email]
[v1]
Tue, 13 Oct 2026 16:55:32 UTC (3,606 KB)
来源：arXiv：cs.AI · arxiv.org