Metraipolis › archive › dc_632cd258d99f

Masked diffusion enables coherent beat tracking — arXiv:2608.04624

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.04624
License (as archived)
CC0-1.0
Archived
2026-10-31T11:14:30.165Z
SHA-256 of the archived text
f7e5334ea4ba260f2da1f2334b878c3977c4456b213afa39f29db0da0e20b84d
Stable URL of this page
https://metraipolis.com/documento/dc_632cd258d99f

How to cite

Masked diffusion enables coherent beat tracking — arXiv:2608.04624 — arXiv (en), https://arxiv.org/abs/2608.04624 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_632cd258d99f (Metraipolis archive dc_632cd258d99f, sha256 f7e5334ea4ba260f2da1f2334b878c3977c4456b213afa39f29db0da0e20b84d)
BibTeX
@misc{dc632cd258d99f,
  title = {Masked diffusion enables coherent beat tracking — arXiv:2608.04624},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.04624},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_632cd258d99f; sha256 f7e5334ea4ba260f2da1f2334b878c3977c4456b213afa39f29db0da0e20b84d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04624. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Current neural networks for beat tracking generate invalid outputs, such as consecutive downbeats and erratic tempo changes, even when these are not present in the training data. Heavy post-processing techniques can alleviate these problems, but the original cause of this inconsistent behaviour remains unknown. We hypothesise that it stems from inadequate modelling of multiple plausible output beat grids, resulting in an invalid mixture of competing interpretations. We propose a masked diffusion approach that properly models multiple outputs and enables the model to build coherent predictions through iterative inference. We devise three modifications to standard masked diffusion that enable its application to beat tracking: independent masking of beats and downbeats during training and inference, a balanced masking scheduler for inference, and peak-picking across inference steps. Our approach reduces erratic behaviours and improves beat-tracking performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.AI).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.04624 — 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 2026-10-31T11:02:13.535Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).