AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

ICR-RL: Deep Reinforcement Learning via In-Context-Regression

David Schiff, Ofir Lindenbaum, Yonathan Efroni

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL · 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
use ICR-RL (TabPFN in-context Q regression) instead of PPO or TRPOICR-RL with TabPFN · CartPole-v1, MountainCar-v0, Acrobot-v1 (250 episodes)reward per episodeImproved
replace latest-episodes context truncation with naive de-duplication once context is full (ablation)ICR-RL · continued learning after context budget is fullreward per episodeImproved
use embeddings de-duplication or reward-variance heuristics instead of naive de-duplication (ablation)ICR-RL context engineering · continued learning after context is fullreward per episodeHurt
use latest-episodes FIFO truncation vs stale context (ablation)ICR-RL context engineering · continued learning after context is fullreward per episodeHurt
use context engineering to shrink context to 512 samples (ablation)ICR-RL · context size studyreward per episode: 512 samples on par with context size 8096No clear effect
apply proper context engineering methodICR-RL · runtimenormalized runtime: nearly 40% reductionImproved
swap TabPFN backbone for TabICL (ablation)ICR-RL · sparse-reward CartPolereward per episodeImproved
raise percentile refit threshold to 0.9 or 0.95 (vs 0.4-0.7) (ablation)ICR-RL · MountainCar-v0 (300 episodes, 5 seeds)convergence: 0.9 and 0.95 nearly identical top performanceImproved
increase context budget B from 256 up to 8192 (ablation)ICR-RL · MountainCar-v0 (300 episodes, 5 seeds)convergenceImproved

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