Metraipolis › archive › dc_3cd062d3e586

Parameter-Efficient Fine-Tuning of Transformer Encoders for Audio-Image-Symbolic Music Retrieval — arXiv:2610.41512

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41512
License (as archived)
CC0-1.0
Archived
2026-11-03T05:46:25.187Z
SHA-256 of the archived text
fd9486eb4e081763ab002df80779d27f32481814ae3cef6be232eebbb9eef6fe
Stable URL of this page
https://metraipolis.com/documento/dc_3cd062d3e586

How to cite

Parameter-Efficient Fine-Tuning of Transformer Encoders for Audio-Image-Symbolic Music Retrieval — arXiv:2610.41512 — arXiv (en), https://arxiv.org/abs/2610.41512 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3cd062d3e586 (Metraipolis archive dc_3cd062d3e586, sha256 fd9486eb4e081763ab002df80779d27f32481814ae3cef6be232eebbb9eef6fe)
BibTeX
@misc{dc3cd062d3e586,
  title = {Parameter-Efficient Fine-Tuning of Transformer Encoders for Audio-Image-Symbolic Music Retrieval — arXiv:2610.41512},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41512},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3cd062d3e586; sha256 fd9486eb4e081763ab002df80779d27f32481814ae3cef6be232eebbb9eef6fe},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41512. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Score-based multimodal music retrieval requires representations that are both discriminative within each modality and compatible across audio recordings, sheet music images, and symbolic scores. This paper investigates the effectiveness of parameter-efficient fine-tuning (PEFT) for adapting pretrained music encoders and examines whether modality-specific improvements translate into better multimodal representations. We compare full fine-tuning, LoRA, and QLoRA across nine pretrained audio, image, and symbolic score encoders on TriScore and the Musical Theme Dataset. Encoders are first evaluated and selected through supervised unimodal classification, using frozen-encoder linear probing as a reference. We then introduce a controlled 2x2 design crossing encoder adaptation with task-specific shared projection, comparing frozen and adapted representations before and after projection. LoRA or QLoRA is selected in 14 of 18 modality-target combinations, frequently matching or outperforming full fine-tuning while optimizing substantially fewer parameters. However, these unimodal gains do not transfer uniformly to multimodal classification. Adaptation and projection are complementary for composer classification on TriScore, largely redundant for TriScore instrument and the Musical Theme Dataset instrument and ensemble tasks, and partially interfering for composer classification on the Musical Theme Dataset. The final adapted-and-projected title representation on TriScore supports piece-level retrieval, reaching average Top-5 and Top-10 accuracies of 55.32 and 64.68, respectively. Cross-modal exclusion and neighborhood analyses further reveal strong modality-specific local structure in the projected spaces. Overall, PEFT provides an effective and computationally efficient alternative to full fine-tuning, but improved unimodal discrimination alone does not guarantee better multimodal organization.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Sound (cs.SD); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2610.41512 — 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 2026-11-03T05:33:32.301Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).