{
 "id": "mtg-upf/HumMusQA",
 "owner": "mtg-upf",
 "name": "HumMusQA",
 "downloads": 58,
 "likes": 0,
 "last_modified": "2026-03-31T11:41:04.000Z",
 "stated_licence": "cc-by-4.0",
 "stated_from": "tag",
 "licence_name": null,
 "licence_link": null,
 "file_licence": null,
 "licence_files": [],
 "sources": [],
 "gated": false,
 "tasks": [
  "question-answering"
 ],
 "citation": "@inproceedings{weck-etal-2026-hummusqa,\n    title = \"{H}um{M}us{QA}: A Human-written Music Understanding {QA} Benchmark Dataset\",\n    author = \"Weck, Benno  and\n      Puentes, Pablo  and\n      Poltronieri, Andrea  and\n      Prabhu, Satyajeet  and\n      Bogdanov, Dmitry\",\n    editor = \"Epure, Elena V.  and\n      Oramas, Sergio  and\n      Doh, SeungHeon  and\n      Ramoneda, Pedro  and\n      Kruspe, Anna  and\n      Sordo, Mohamed\",\n    booktitle = \"Proceedings of the 4th Workshop on {NLP} for Music and Audio ({NLP}4{M}us{A} 2026)\",\n    month = mar,\n    year = \"2026\",\n    address = \"Rabat, Morocco\",\n    publisher = \"Association for Computational Linguistics\",\n    url = \"https://aclanthology.org/2026.nlp4musa-1.9/\",\n    doi = \"10.18653/v1/2026.nlp4musa-1.9\",\n    pages = \"58--67\",\n    ISBN = \"979-8-89176-369-2\",\n    abstract = \"The evaluation of music understanding in Large Audio-Language Models (LALMs) requires a rigorously defined benchmark that truly tests whether models can perceive and interpret music, a standard that current data methodologies frequently fail to meet.This paper introduces a meticulously structured approach to music evaluation, proposing a new dataset of 320 hand-written questions curated and validated by experts with musical training, arguing that such focused, manual curation is superior for probing complex audio comprehension.To demonstrate the use of the dataset, we benchmark six state-of-the-art LALMs and additionally test their robustness to uni-modal sh",
 "signal": "permissive",
 "reason": "The declared licence (cc-by-4.0) is classified as permissive.",
 "revision": "ca7bad071bde0b27e55eca050f2e88b49ae01f56",
 "stated_evidence": "https://huggingface.co/datasets/mtg-upf/HumMusQA/blob/ca7bad071bde0b27e55eca050f2e88b49ae01f56/README.md",
 "file_evidence": "https://huggingface.co/datasets/mtg-upf/HumMusQA/tree/ca7bad071bde0b27e55eca050f2e88b49ae01f56",
 "card_nc": false
}