Stanford Research: Robot Learns New Tasks via Single Demo with BPP
Haoyu_Xiong_ · x · 2026-08-27
Researchers from Stanford and UC Berkeley released Behavior Prompting Policy (BPP), enabling robots to perform new tasks at inference time using a single human demonstration without fine-tuning.
Core Contributions:
- BPP Architecture: An in-context visuomotor policy translating behavior prompts (demo sequences) and observations into actions.
- iPhUMI: A handheld manipulation interface for collecting diverse behavior prompts.
Key Findings:
- Task diversity drives prompting capability more than the number of demos per task.
- Behavior prompts act as dense sub-goal conditioning.
Open Source & Eval:
- Released BPP code, iPhUMI code, and simulation benchmarks.
- Introduced DrawAnything and LIBERO-Gen benchmarks to evaluate test-time adaptation.
More from Embodied
- Neuralink demos patient creating digital art with mind — beffjezos · 2026-08-27
- Funny Robot Moments: A Look at AI's Humorous Side — LinusEkenstam · 2026-08-27
- DEEP Robotics unveils 180cm humanoid DR02 at AI Expo — carlosdponx · 2026-08-27
- Ray Summit 2026: Advances in robotics sim and vision models — AndrewDai · 2026-08-27
- Humanoid 100m Sprint Record Broken Again: 8.6 Seconds in 3 Days — LinusEkenstam · 2026-08-27
- New Podcast Episode: Coupled Local and Global World Models in Robotics — micoolcho · 2026-08-27