AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention

Ruben Solozabal, Velibor Bojkovic, Hilal AlQuabeh, Klea Ziu, Kentaro Inui, Martin Takáč

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models · 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
replace Legendre polynomial bases with wavelet frames in S4S4 / WaveSSM · PTB-XL overall AUROCAUROC: 0.934 -> 0.942Improved
replace Legendre bases with wavelet framesS4 / WaveSSM · Speech Commands SC35 raw 16kHzaccuracy (%): 96.08 -> 96.55Improved
test wavelet SSM under zero-shot downsampling to 8kHzWaveSSM · Speech Commands SC35 8kHzaccuracy (%): 91.59 (S4 FouT) vs 90.14 (Morlet scaled)Hurt
replace Legendre bases with wavelet framesS4 / WaveSSM · Long Range Arena (ListOps, Text, Retrieval, Image)accuracy (%): ListOps 59.60 -> 61.79Improved
replace Legendre bases with wavelet framesS4 / WaveSSM · LRA PathXaccuracy (%): 96.35 -> infeasibleHurt
use Gauss, Mexhat or dpss frames with scaled measure (ablation)WaveSSM · LRA Pathfinderaccuracy (%): 94.20 -> infeasibleHurt
use translated measure instead of scaled measure for Morlet (ablation)WaveSSM Morlet · LRA Pathfinderaccuracy (%): 94.94 -> 73.3Hurt
replace Legendre with wavelet frames in S4S4 / WaveSSM · Informer univariate forecasting MSEMSE: ECL 0.432 -> 0.268Improved
add frame tightening (whitening of the frame) to wavelet frames (ablation)WaveSSM frames · condition number of frame operatorcondition numberImproved
use wavelet frames instead of Legendre polynomials for window copyingWaveSSM · Continuous Window Copyingreconstruction MSEImproved
add input-dependent selection (Mamba) instead of linear S4S4 · Continuous Window Copyingreconstruction MSEImproved

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