Metraipolis › archive › dc_dc6fff9af965
Don't CLAP: Are Music-Text Models Bag-of-Words? — arXiv:2609.30540
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30540
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T02:04:13.603Z
- SHA-256 of the archived text
be6659dabafc9a4b2cd2574d9760f0e20fc49f64bbc1f96bee7c4670a8b3209c- Stable URL of this page
https://metraipolis.com/documento/dc_dc6fff9af965
How to cite
Don't CLAP: Are Music-Text Models Bag-of-Words? — arXiv:2609.30540 — arXiv (en), https://arxiv.org/abs/2609.30540 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dc6fff9af965 (Metraipolis archive dc_dc6fff9af965, sha256 be6659dabafc9a4b2cd2574d9760f0e20fc49f64bbc1f96bee7c4670a8b3209c)
BibTeX
@misc{dcdc6fff9af965,
title = {Don't CLAP: Are Music-Text Models Bag-of-Words? — arXiv:2609.30540},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30540},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dc6fff9af965; sha256 be6659dabafc9a4b2cd2574d9760f0e20fc49f64bbc1f96bee7c4670a8b3209c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30540. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Text-to-music systems are assessed on audio quality and on how faithfully the music follows its prompt, and the CLAP score, the cosine similarity between a music-text model's audio and text embeddings, is the standard objective metric of faithfulness. We ask how accurately that score reflects the text: when an attribute is linked to an instrument (e.g., distorted guitar), does the text embedding capture that binding? To find out, we introduce an attribute swap perturbation: the caption of a real recording is edited by exchanging exactly one property, timbre, lead versus accompaniment, or order of first appearance, between two instruments. We then test four contrastive music-text models and one large audio-language model on whether the audio scores higher against the original caption than against the perturbed one. No contrastive model distinguishes the two captions reliably. The audio-language model does better, but further experiments show that its advantage rests largely on audio-agnostic language priors. Our results thus provide compelling evidence that the CLAP score and related metrics do not capture fine-grained musical meaning or attribute bindings; their representation is closer to a bag-of-words that leaves them insensitive to meaning-changing perturbations of the caption. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.CL). Categorie arXiv: Sound (cs.SD) ; Computation and Language (cs.CL); Audio and Speech Processing (eess.AS). Pagina del lavoro: https://arxiv.org/abs/2609.30540 — 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 2026-10-08T01:38:49.477Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).