OmniScore-Bench: Evaluating Vision-Language Models on Musical Score Reading
arXiv:2611.17870v1 Announce Type: new Abstract: Vision-language models (VLMs) can answer questions about musical scores, but their ability to recognise notation, recover its structure and interpret its musical meaning requires separate evaluation. Existing optical music recognition (OMR) resources establish detection and transcription measurements; a transcription score alone leaves the strengths and bottlenecks of a VLM score reader unresolved. We introduce OmniScore-Bench, an evaluation of OMR
Read original sourceSource record
Project analysis
Project analysis is not yet available. See the source record above.