Metraipolis › archive › dc_bf669bf8d1bf

ResMamba: A Dual-Path Convolution-Mamba Model for Music Genre Classification — arXiv:2611.00746

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.00746
License (as archived): CC0-1.0
Archived: 2026-11-03T10:34:24.767Z
SHA-256 of the archived text: e0b1e42e238b78f9e7cbd2b63e58e7e4c8a5b5929269e69e1581ab3ba25aeb13
Stable URL of this page: https://metraipolis.com/documento/dc_bf669bf8d1bf

How to cite

ResMamba: A Dual-Path Convolution-Mamba Model for Music Genre Classification — arXiv:2611.00746 — arXiv (en), https://arxiv.org/abs/2611.00746 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf669bf8d1bf (Metraipolis archive dc_bf669bf8d1bf, sha256 e0b1e42e238b78f9e7cbd2b63e58e7e4c8a5b5929269e69e1581ab3ba25aeb13)

BibTeX

@misc{dcbf669bf8d1bf,
  title = {ResMamba: A Dual-Path Convolution-Mamba Model for Music Genre Classification — arXiv:2611.00746},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00746},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf669bf8d1bf; sha256 e0b1e42e238b78f9e7cbd2b63e58e7e4c8a5b5929269e69e1581ab3ba25aeb13},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00746. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Music genre classification remains a fundamental yet challenging task in music information retrieval, requiring models that capture local time--frequency patterns and long-range temporal dependencies. To address this, we propose ResMamba, a hybrid architecture that augments a ResNet-style four-stage backbone with Mamba blocks. Building upon the selective structured state space sequence model, ResMamba first adapts the selective SSM into a 4D-compatible residual Mamba block that captures global dependencies at linear complexity. We then design the ConvMamba dual-path block, which combines this adapted Mamba branch with a parallel convolutional branch through a learned $1\times1$ fusion, jointly modeling global sequential context and local spectro-temporal patterns. Furthermore, we adopt a multi-scale fusion head that aggregates hierarchical representations across all four stages. Experiments on the GTZAN dataset demonstrate that our full ResMamba model achieves a 5-fold cross-validation accuracy of $88.99\% \pm 1.58\%$, surpassing the strongest published method in our comparison (88.64\%). All models are trained from scratch without any pretraining, external data, or data augmentation. Beyond GTZAN, ResMamba also generalizes to the larger FMA benchmarks under the same training protocol, attaining $64.52\% \pm 1.88\%$ on FMA-Small under 10-fold cross-validation --- the best published from-scratch result --- and $65.75\%$ on FMA-Medium under the official artist-level split, surpassing the widely cited SVM baseline of Defferrard \textit{et al.} (63\%).

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2611.00746 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T10:29:48.721Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).