AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

ELMUR: External Layer Memory with Update/Rewrite for Long-Horizon RL

Egor Cherepanov, Alexey Kovalev, Aleksandr Panov

ICLR 2026 · Reinforcement learning › Visual and representation RL · approach: In-context and adaptive RL

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 LRU memory update instead of removing it (ablation)ELMUR · MIKASA-Robo RememberColor3-v0normalized score: 0.43 -> 1.00Improved
use layer-local memory instead of memory shared across layers (ablation)ELMUR · MIKASA-Robo RememberColor3-v0normalized score: 0.45 -> 1.00Improved
add relative bias to token-memory cross-attention (ablation)ELMUR · MIKASA-Robo RememberColor3-v0normalized score: 0.95 -> 1.00Improved
add relative bias and LRU together (ablation)ELMUR · MIKASA-Robo RememberColor3-v0normalized score: 0.22 -> 1.00Improved
replace MoE-FFN with MLP-FFN (ablation)ELMUR · MIKASA-Robo RememberColor3-v0normalized score: 1.00 -> 1.00No clear effect
memory slots M at least the number of needed segments N (ablation)ELMUR · MIKASA-Robo RememberColor3-v0success rateImproved
intermediate LRU blending lambda about 0.4 to 0.6 (ablation)ELMUR · MIKASA-Robo RememberColor3-v0success rateHurt
larger memory initialization scale sigma (ablation)ELMUR · MIKASA-Robo RememberColor3-v0success rateImproved
external layer memory with LRU instead of RATE on TakeItBackELMUR · MIKASA-Robo TakeItBack-v0success rate: 0.42 -> 0.78Improved
external layer memory with LRU instead of RATE on POPGym puzzlesELMUR · POPGym puzzle tasks (33)aggregated return: 0.45 -> 1.2Improved
external memory on reactive POPGym tasksELMUR · POPGym reactive tasks (15)aggregated return: 9.3 (DT) vs 9.2No clear effect
condition ELMUR on returns-to-go and actions rather than observations only (ablation)ELMUR · D4RL MuJoConormalized score: average 76.4, about 2% higherNo clear effect

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