WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917294849916928 |
|---|---|
| author | Solozabal, Ruben Bojkovic, Velibor Alquabeh, Hilal Ziu, Klea Inui, Kentaro Takac, Martin |
| author_facet | Solozabal, Ruben Bojkovic, Velibor Alquabeh, Hilal Ziu, Klea Inui, Kentaro Takac, Martin |
| contents | State-space models (SSMs) have emerged as a powerful foundation for long-range sequence modeling, with the HiPPO framework showing that continuous-time projection operators can be used to derive stable, memory-efficient dynamical systems that encode the past history of the input signal. However, existing projection-based SSMs often rely on polynomial bases with global temporal support, whose inductive biases are poorly matched to signals exhibiting localized or transient structure. In this work, we introduce \emph{WaveSSM}, a collection of SSMs constructed over wavelet frames. Our key observation is that wavelet frames yield a localized support on the temporal dimension, useful for tasks requiring precise localization. Empirically, we show that on equal conditions, \textit{WaveSSM} outperforms orthogonal counterparts as S4 on real-world datasets with transient dynamics, including physiological signals on the PTB-XL dataset and raw audio on Speech Commands. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_22266 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention Solozabal, Ruben Bojkovic, Velibor Alquabeh, Hilal Ziu, Klea Inui, Kentaro Takac, Martin Machine Learning Sound State-space models (SSMs) have emerged as a powerful foundation for long-range sequence modeling, with the HiPPO framework showing that continuous-time projection operators can be used to derive stable, memory-efficient dynamical systems that encode the past history of the input signal. However, existing projection-based SSMs often rely on polynomial bases with global temporal support, whose inductive biases are poorly matched to signals exhibiting localized or transient structure. In this work, we introduce \emph{WaveSSM}, a collection of SSMs constructed over wavelet frames. Our key observation is that wavelet frames yield a localized support on the temporal dimension, useful for tasks requiring precise localization. Empirically, we show that on equal conditions, \textit{WaveSSM} outperforms orthogonal counterparts as S4 on real-world datasets with transient dynamics, including physiological signals on the PTB-XL dataset and raw audio on Speech Commands. |
| title | WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention |
| topic | Machine Learning Sound |
| url | https://arxiv.org/abs/2602.22266 |