Research map / Reinforcement learning / Imitation learning
Translating Flow to Policy via Hindsight Online Imitation
Yitian Zheng, Zhangchen Ye, Weijun Dong, Shengjie Wang, Yuyang Liu, Chongjie Zhang, Chuan Wen, Yang Gao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add hindsight online imitation to the base policy | flow-conditioned policy · 7 manipulation tasks | success rate: over 2x base policy; avg 84.0% | Improved |
| use flow-guided HinFlow instead of Online VPT | low-level policy · 7 manipulation tasks | success rate | Improved |
| online interaction for 10,000 steps | HinFlow · Real-world mouse task | success rate: 40% -> 95% | Improved |
| add cross-embodiment action-free videos to planner training | HinFlow · Place Book and Poke Cube | success rate: gain of over 40 percentage points | Improved |
| use flow conditioning instead of BC for zero-shot distractors | low-level policy · Place Butter with extra distractors | success rate: 0.0 -> 92.8 | Improved |
| use flow conditioning instead of BC for unseen target object | low-level policy · Place Butter, unseen object | success rate: 6.5 -> 96.2 | Improved |
| pretrain with 0 action-labeled demos (ablation) | HinFlow · LIBERO / ManiSkill | success rate | Hurt |
| vary number of pretrain demos (1-10) before online imitation (ablation) | HinFlow · LIBERO / ManiSkill | final success rate | No clear effect |
| flow length 8, 12 or 16 instead of 4 (ablation) | HinFlow · LIBERO / ManiSkill | success rate | Improved |
Closest papers by meaning
- ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- FlowPolicy: Enabling Fast and Robust 3D Flow-Based Policy via Consistency Flow Matching for Robot Manipulation (AAAI 2025 · similarity 0.89)
- Zero-Shot Robotic Manipulation with Pre-Trained Image-Editing Diffusion Models (ICLR 2024 · similarity 0.88)
- Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry (NeurIPS 2026 · similarity 0.88)
- GO-DICE: Goal-Conditioned Option-Aware Offline Imitation Learning via Stationary Distribution Correction Estimation (AAAI 2024 · similarity 0.88)
- Efficient Robotic Policy Learning via Latent Space Backward Planning (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
