AI Paper Atlas

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

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Dexterous manipulation transfer

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 changedWhereMeasuredEffect
add human-in-the-loop interventions (ablation)FTC · three contact-rich taskssuccess rate: 0 -> 100Improved
add affordance-guided reward (ablation)FTC · three-task averagetraining time: 49.7 -> 33.3Improved
add residual RL (ablation)FTC · three-task averagetraining time: 56 -> 33.3Improved
add residual RL and affordance reward together (ablation)FTC · USB insertion (hard)success rate: 0 -> 100Improved
add residual RL and affordance reward together (ablation)FTC · three-task averagesuccess rate: 66.7 -> 100Improved
add affordance reward alone (ablation)FTC · three-task averagesuccess rate: 100 -> 100No clear effect
add residual RL (ablation)FTC · OOD USB insertion (hard)average success rate: 11.3 -> 86.3Improved
use FTC instead of HIL-SERLFTC · USB insertion (hard)success rate: 0 -> 100Improved
use FTC instead of HIL-SERLFTC · six-task averagetraining time: 56.7 -> 27.5Improved
use FTC instead of HIL-SERLFTC · six-task averagesuccess rate: 82 -> 100Improved
use FTC instead of HIL-SERL in OOD positionsFTC · USB insertion (hard) Pos 5-8success rate: 0 -> 90Improved
strong lighting conditionFTC · Charger grasp-insertionsuccess rate: 100 -> 0Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy