AI Paper Atlas

Research map / Reinforcement learning / Offline RL

RAD: Retrieval High-quality Demonstrations to Enhance Decision-making

Lu Guo, Yixiang Shan, Zhengbang Zhu, Qifan Liang, Lichang Song, Ting Long, Weinan Zhang, Yi Chang

ICML 2026 · Reinforcement learning › Offline RL · approach: Offline-to-online fine-tuning · 0 citations (OpenAlex)

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 retrieval of high-return target states, step estimation and planning to DiffuserLiteDiffuserLite · AntMaze (average)normalized score: 77.6 -> 83.3Improved
add retrieval-augmented targets to DiffuserLiteDiffuserLite · Gym-MuJoCo locomotion (average)normalized score: 86.0 -> 87.5Improved
add retrieval-augmented targets to DiffuserDiffuser · Maze2D (average)normalized score: 119.5 -> 134.9Improved
add retrieval-augmented targets to DiffuserLiteDiffuserLite · FrankaKitchen (average)normalized score: 74.0 -> 72.1Hurt
add retrieval-augmented targets to DiffuserDiffuser · Medium-Expert locomotionnormalized scoreNo clear effect
add retrieval-augmented targets to DiffuserLite under region-removed shiftDiffuserLite · AntMaze-Medium-Play, 0% coveragenormalized score: 0.0 -> 36.7Improved
add retrieval-augmented targets to DiffuserLite under 20% coverageDiffuserLite · AntMaze-Large-Play, 20% coveragenormalized score: 0.0 -> 47.8Improved
add retrieval-augmented targets to DiffuserLite under 0% coverageDiffuserLite · AntMaze-Large-Play, 0% coveragenormalized score: 0.0 -> 0.0No clear effect
add retrieval targets to DiffuserLite under shifted initial statesDiffuserLite · Hopper, Random initial statesnormalized score: 60.4 -> 90.2Improved
add retrieval targets vs MOPO under shifted initial statesDL-RAD · HalfCheetah, Random initial statesnormalized score: 62.3 -> 39.2Hurt
remove target selection module (-TS) (ablation)D-RAD · Medium-Expert/Medium/Medium-Replay MuJoConormalized scoreImproved
replace retrieved targets with random targets (-TSR) (ablation)D-RAD · MuJoCo locomotionnormalized scoreImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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