Annotation-Free One-Shot Imitation Learning for Multi-Step Manipulation Tasks
On the map: Offline RL and imitation learning from data → Robot manipulation imitation learning
Abstract (arXiv)
Recent advances in one-shot imitation learning have enabled robots to acquire new manipulation skills from a single human demonstration. While existing methods achieve strong performance on single-step tasks, they remain limited in their ability to handle long-horizon, multi-step tasks without additional model training or manual annotation. We propose a method that can be applied to this setting provided a single demonstration without additional model training or manual annotation. We evaluated our method on multi-step and single-step manipulation tasks where our method achieves an average success rate of 82.5% and 90%, respectively. Our method matches and exceeds the performance of the baselines in both these cases. We also compare the performance and computational efficiency of alternative pre-trained feature extractors within our framework.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 24 of its 26 references that are in the map. First posted on arXiv 29 Sep 2025.
Builds on
24 of 26 references.
Closest ancestors · 13
Same direction, newest first
- 2025OSVI-WM: One-Shot Visual Imitation for Unseen Tasks using World-Model-Guided Trajectory Generation26 May
- 2025One-Shot Dual-Arm Imitation Learning10 Mar
- 2025FUNCTO: Function-Centric One-Shot Imitation Learning for Tool Manipulation17 Feb
- 2024Instant Policy: In-Context Imitation Learning via Graph Diffusion19 Nov
- 2024Keypoint Abstraction using Large Models for Object-Relative Imitation Learning30 Oct
- 2024ALOHA Unleashed: A Simple Recipe for Robot Dexterity17 Oct
- 2024One-Shot Imitation Learning with Invariance Matching for Robotic Manipulation21 May
- 2024Keypoint Action Tokens Enable In-Context Imitation Learning in Robotics28 Mar
- 2024DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Models20 Feb
- 2023One-Shot Imitation Learning: A Pose Estimation Perspective18 Oct
- 2023One-shot Imitation Learning via Interaction Warping21 Jun
- 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware23 Apr
1 more
Other references · 11
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
- 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
- 2024SAM 2: Segment Anything in Images and Videoscited by 2,784
- 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
- 2023LightGlue: Local Feature Matching at Light Speedcited by 583
- 2025Perception Encoder: The best visual embeddings are not at the output of the networkcited by 315
- 2023RVT: Robotic View Transformer for 3D Object Manipulationcited by 224
- 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
- 2024CleanDIFT: Diffusion Features without Noisecited by 42
- 2024One-Shot Imitation under Mismatched Executioncited by 11
- 2025Elastic Motion Policy: An Adaptive Dynamical System for Robust and Efficient One-Shot Imitation Learningcited by 4
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2021Towards More Generalizable One-shot Visual Imitation Learning
- 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration
- 2023RH20T: A Comprehensive Robotic Dataset for Learning Diverse Skills in One-Shot
- 2023Waypoint-Based Imitation Learning for Robotic Manipulation
- 2025Learning a Thousand Tasks in a Day
- 2022Eliciting Compatible Demonstrations for Multi-Human Imitation Learning
- 2025Robotic Manipulation via Imitation Learning: Taxonomy, Evolution, Benchmark, and Challenges
- 2024VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections