Research map / Reinforcement learning / Visual and representation RL
Focus-Then-Contact: Speeding Up Robotic Contact-Rich Task Learning with Affordance-Guided Real-World Residual Reinforcement Learning
Guanren Qiao, Ruixiang Ouyang, Sheng Xu, Ruixing Jin, Yueci Deng, Yunxin Tai, Kui Jia, Guiliang Liu
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 human-in-the-loop interventions (ablation) | FTC · three contact-rich tasks | success rate: 0 -> 100 | Improved |
| add affordance-guided reward (ablation) | FTC · three-task average | training time: 49.7 -> 33.3 | Improved |
| add residual RL (ablation) | FTC · three-task average | training time: 56 -> 33.3 | Improved |
| add residual RL and affordance reward together (ablation) | FTC · USB insertion (hard) | success rate: 0 -> 100 | Improved |
| add residual RL and affordance reward together (ablation) | FTC · three-task average | success rate: 66.7 -> 100 | Improved |
| add affordance reward alone (ablation) | FTC · three-task average | success rate: 100 -> 100 | No clear effect |
| add residual RL (ablation) | FTC · OOD USB insertion (hard) | average success rate: 11.3 -> 86.3 | Improved |
| use FTC instead of HIL-SERL | FTC · USB insertion (hard) | success rate: 0 -> 100 | Improved |
| use FTC instead of HIL-SERL | FTC · six-task average | training time: 56.7 -> 27.5 | Improved |
| use FTC instead of HIL-SERL | FTC · six-task average | success rate: 82 -> 100 | Improved |
| use FTC instead of HIL-SERL in OOD positions | FTC · USB insertion (hard) Pos 5-8 | success rate: 0 -> 90 | Improved |
| strong lighting condition | FTC · Charger grasp-insertion | success rate: 100 -> 0 | Hurt |
Closest papers by meaning
- Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes (AAAI 2025 · similarity 0.87)
- LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning (NeurIPS 2026 · similarity 0.86)
- Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning (ICML 2025 · similarity 0.86)
- Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections (NeurIPS 2025 · similarity 0.86)
- Plan-Seq-Learn: Language Model Guided RL for Solving Long Horizon Robotics Tasks (ICLR 2024 · similarity 0.85)
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.85)
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.
