{"ID":2862935,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.26521","arxiv_id":"2509.26521","title":"MUSE-Explainer: Counterfactual Explanations for Symbolic Music Graph Classification Models","abstract":"Interpretability is essential for deploying deep learning models in symbolic music analysis, yet most research emphasizes model performance over explanation. To address this, we introduce MUSE-Explainer, a new method that helps reveal how music Graph Neural Network models make decisions by providing clear, human-friendly explanations. Our approach generates counterfactual explanations by making small, meaningful changes to musical score graphs that alter a model's prediction while ensuring the results remain musically coherent. Unlike existing methods, MUSE-Explainer tailors its explanations to the structure of musical data and avoids unrealistic or confusing outputs. We evaluate our method on a music analysis task and show it offers intuitive insights that can be visualized with standard music tools such as Verovio.","short_abstract":"Interpretability is essential for deploying deep learning models in symbolic music analysis, yet most research emphasizes model performance over explanation. To address this, we introduce MUSE-Explainer, a new method that helps reveal how music Graph Neural Network models make decisions by providing clear, human-friend...","url_abs":"https://arxiv.org/abs/2509.26521","url_pdf":"https://arxiv.org/pdf/2509.26521v1","authors":"[\"Baptiste Hilaire\",\"Emmanouil Karystinaios\",\"Gerhard Widmer\"]","published":"2025-09-30T16:58:07Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\"]","methods":"[\"Graph Neural Network\"]","has_code":false}