Structured Sparse Transition Matrices to Enable State Tracking in State-Space Models

NeurIPSSpotlight2025

Authors
Aleksandar Terzic, Nicolas Menet, Michael Hersche, Thomas Hofmann, Abbas Rahimi
Venue
NeurIPS 2025
Track
Spotlight

TL;DR

Modern state-space models (SSMs) often utilize structured transition matrices which enable efficient computation but pose restrictions on the model’s expressivity, as measured in terms of the ability…

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

efficient

← All NeurIPS 2025 Spotlight papers · Browse the whole archive