# Direct content-based retrieval from music scores images

arXiv: https://arxiv.org/abs/2605.22255 · HTML: https://mlsanity.com/paper/2605.22255/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 57 references that are in the map.
- First posted on arXiv 21 May 2026.

## Builds on (most cited first, 9 of 9)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Hard Negative Mixing for Contrastive Learning](https://mlsanity.com/paper/2010.01028/) (2020)
- [PaliGemma 2: A Family of Versatile VLMs for Transfer](https://mlsanity.com/paper/2412.03555/) (2024)
- [Understanding Optical Music Recognition](https://mlsanity.com/paper/1908.03608/) (2019)
- [Sheet Music Transformer: End-To-End Optical Music Recognition Beyond Monophonic Transcription](https://mlsanity.com/paper/2402.07596/) (2024)
- [End-to-End Full-Page Optical Music Recognition for Pianoform Sheet Music](https://mlsanity.com/paper/2405.12105/) (2024)
- [NOTA: Multimodal Music Notation Understanding for Visual Large Language Model](https://mlsanity.com/paper/2502.14893/) (2025)
- [MusiXQA: Advancing Visual Music Understanding in Multimodal Large Language Models](https://mlsanity.com/paper/2506.23009/) (2025)
- [Combining piano performance dimensions for score difficulty classification](https://mlsanity.com/paper/2306.08480/) (2023)

## Built on by (most cited first, 1 of 1)

- [DEMUN: Fast and accurate discovery of music notation in very large collections](https://mlsanity.com/paper/2606.31956/) (2026)

## Abstract (arXiv)

The digitization of musical scores plays a crucial role in their preservation and accessibility, yet information retrieval still depends mainly on metadata searches, such as by title or composer. Content based search in music score images remains underexplored compared to text documents, despite its potential value for musicians, musicologists, and educators. This work contributes to the field by first studying which characteristics of a score are most relevant for search and by defining a systematic method to build query datasets from any annotated corpus. We also consider diverse methods for content-based search on music score images, ranging from transcription-based approaches relying on Optical Music Recognition (OMR), to a transcription-free Transformer model trained to recognize queries directly from score images, and a text-prompted Large Language Model. Our experiments evaluate these models on four corpora exhibiting diverse characteristics in terms of dataset size, image quality, and typesetting mechanisms. Overall, each method excels under different conditions: OMR-based pipelines achieve higher in-domain retrieval, whereas transcription-free models handle domain variability more effectively.