AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

AIRE-Prune: Asymptotic Impulse-Response Energy for State Pruning in State Space Models

Apurba Prasad Padhy, Fernando F. Camacho, Saibal Mukhopadhyay

ICLR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Structured network pruning · 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
score states by asymptotic impulse-response energy with prefix normalization instead of LASTS5 · LRA averageaverage accuracy drop (pp) / avg pruning: 0.52 -> 0.29 pp drop; 60.8% vs 33% pruningImproved
use AIRE-Prune instead of Global H-infS5 · Pathfinder (30% pruned baseline)accuracy (%): 57.20 -> 95.15Improved
use AIRE-Prune instead of LAST on RetrievalS5 · Retrievalaccuracy (%): 90.42 -> 90.11Hurt
prune ListOps with AIRE-PruneS5 · ListOpsaccuracy (%): 61.48 -> 61.48 at 20% pruningNo clear effect
apply AIRE-Prune to S4DS4D · Textaccuracy (%): 86.40 -> 85.83 at 90% prunedMixed
apply AIRE-Prune to MambaMamba · Imageaccuracy (%): 69.82 -> 68.97 at 80% prunedMixed
prune S5 states with AIRE-PruneS5 · LRA Pathfinderinference speedup: 2.82x speedup, 64.7% fewer parametersImproved
prune states in S5 by AIRE-Prune on Speech CommandsS5 · Speech Commandsaccuracy (%): 96.43 -> 96.40 at 45% prunedNo 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