Metraipolis › archive › dc_cfdbe17ae30b

CoJEPA: Combining Contrastive Learning and JEPA for Global-Local Music Representations — arXiv:2608.30974

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30974
License (as archived)
CC0-1.0
Archived
2026-10-17T07:55:10.110Z
SHA-256 of the archived text
4a4ce6db78be797d82ccca6acd75eca80ae07be8a46477150876bad26685cb29
Stable URL of this page
https://metraipolis.com/documento/dc_cfdbe17ae30b

How to cite

CoJEPA: Combining Contrastive Learning and JEPA for Global-Local Music Representations — arXiv:2608.30974 — arXiv (en), https://arxiv.org/abs/2608.30974 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cfdbe17ae30b (Metraipolis archive dc_cfdbe17ae30b, sha256 4a4ce6db78be797d82ccca6acd75eca80ae07be8a46477150876bad26685cb29)
BibTeX
@misc{dccfdbe17ae30b,
  title = {CoJEPA: Combining Contrastive Learning and JEPA for Global-Local Music Representations — arXiv:2608.30974},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30974},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cfdbe17ae30b; sha256 4a4ce6db78be797d82ccca6acd75eca80ae07be8a46477150876bad26685cb29},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30974. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Joint-Embedding Predictive Architecture (JEPA) has shown strong performance in learning rich representations through self-supervised prediction in latent space. However, it typically relies on teacher--student architecture with an EMA to stabilise training, and can tend to yield uninformative representations. Contrastive learning is stable to train and produces strong global representations, but remains limited on local tasks by the global nature of its objective. In this work, we combine both into CoJEPA: a single shared backbone jointly trained with a JEPA objective on masked sequence tokens and a contrastive objective on the class token. The contrastive gradient provides stability, removing the need for an EMA teacher entirely, while JEPA enriches the sequence tokens via local predictions that contrastive learning alone cannot provide. Crucially, no extra parameters are added to the backbone: the same model is guided towards richer representations purely through the design of its training signal. CoJEPA takes the best of both worlds, outperforming or matching both individual methods across global and local MIR tasks, with a particularly strong advantage on tonal and harmonic understanding, and without any task-specific architectural changes. CoJEPA shows that combining objectives with complementary inductive biases can substitute for scale, encouraging future work to invest in smarter training objectives over ever-larger models.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS); Signal Processing (eess.SP).
Pagina del lavoro: https://arxiv.org/abs/2608.30974 — 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 2026-10-17T07:28:26.932Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).