Research map / Reinforcement learning / Imitation learning
RoboCLIP: One Demonstration is Enough to Learn Robot Policies
Sumedh Anand Sontakke, Jesse Zhang, Séb Arnold, Karl Pertsch, Erdem Biyik, Dorsa Sadigh, Chelsea Finn, Laurent Itti
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 |
|---|---|---|---|
| finetune the RoboCLIP-pretrained policy with a single demonstration of observations and actions | RoboCLIP · Coffee-Push and Faucet-Open (Metaworld) | reward: ~200% | Improved |
| use more than one video demonstration for reward generation (ablation) | RoboCLIP · in-domain demonstration setup | downstream reward | Hurt |
| replace S3D video-language model with single-image CLIP using the final frame (ablation) | RoboCLIP · language-conditioned pretraining | zero-shot performance | Hurt |
| use a randomly initialized S3D instead of a pretrained one (ablation) | RoboCLIP · in-domain video demonstration setup | pretraining reward signal | Hurt |
| edit a button-pushing video embedding by subtracting 'button' text and adding 'drawer' text | RoboCLIP · Drawer-Close | zero-shot score | Improved |
| use RoboCLIP language reward instead of GAIL and AIRL baselines | pretraining reward · Drawer-Close, Door-Close, Button-Press | true task reward | Improved |
| use RoboCLIP out-of-domain video reward instead of GAIL and AIRL | pretraining reward · Button-Press, Drawer-Open, Door-Open | zero-shot task performance: 2-3x | Improved |
Closest papers by meaning
- Expert Proximity as Surrogate Rewards for Single Demonstration Imitation Learning (ICML 2024 · similarity 0.85)
- Subtask-Aware Visual Reward Learning from Segmented Demonstrations (ICLR 2025 · similarity 0.85)
- Imitation Learning from Observation with Automatic Discount Scheduling (ICLR 2024 · similarity 0.84)
- RLIF: Interactive Imitation Learning as Reinforcement Learning (ICLR 2024 · similarity 0.84)
- Deep Demonstration Tracing: Learning Generalizable Imitator Policy for Runtime Imitation from a Single Demonstration (ICML 2024 · similarity 0.84)
- Learning from Demonstrations via Capability-Aware Goal Sampling (NeurIPS 2025 · similarity 0.84)
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.
