AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

RMAAT: Astrocyte-Inspired Memory Compression and Replay for Efficient Long-Context Transformers

Md Zesun Ahmed Mia, Malyaban Bal, Abhronil Sengupta

ICLR 2026 · Architectures and efficiency › Attention and transformers · approach: Linear attention approximations · 1 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 memory retention factor (ablation)RMAAT · LRA Retrieval (8K)accuracy (%): 80.5 -> 83.2Improved
add memory retention factor (ablation)RMAAT · LRA Text (4K)accuracy (%): 64.9 -> 65.9Improved
train with AMRB instead of BPTT (ablation)RMAAT · LRA Retrieval (8K)peak memory (GB): 15.0 -> 3.4Improved
train with AMRB instead of BPTT (ablation)RMAAT · LRA Text (4K)peak memory (GB): 22.0 -> 5.1Improved
add retention factor and AMRB to RLT (ablation)RLT · LRA Retrieval (8K)accuracy (%): 78.4 -> 79.2Improved
add astromorphic attention (H_astro, P) on top of RLT+AMRB (ablation)RLT + AMRB · LRA Retrieval (8K)accuracy (%): 79.2 -> 83.2Improved
process longer total context (more segments) (ablation)RMAAT · LRA Retrievalaccuracy (%): 65.3 -> 71.5 -> 83.2Improved
RMAAT recurrence/memory vs non-recurrent ATAT · LRA averageaccuracy (%): 56.4 -> 68.0Improved
RMAAT vs RMT with standard attention and BPTTRMT · LRA Retrievaltraining speed: 1x -> 1.73xImproved
RMAAT vs RMT on ListOpsRMT · LRA ListOps (2K)accuracy (%): 37.4 -> 38.9No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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