aiqbench analysis
This research demonstrates how pre-trained symbolic music transformers can encode distinct jazz pianist identities, enabling high-accuracy style transfer via cross-attention conditioning. The study validates that these models not only replicate specific artistic voices but also identify characteristic musical gestures within performances with significant precision. By leveraging synthetic data to train classifiers, the authors successfully isolate the latent features that define individual stylistic signatures in complex musical compositions.
Learning Jazz Pianist Style with Cross-Attention Conditioning
arXiv:2610.02918v1 Announce Type: cross Abstract: Jazz pianists develop distinctive traits that experienced listeners can often identify within seconds, yet the features underlying this recognition resist formal description. We study jazz pianist style through the lens of a pretrained symbolic music transformer, showing that its learned representations already encode pianist identity well enough for highly accurate classification across two benchmarks. We then augment the transformer with cross-attention over learned pianist identity embeddings, enabling it to generate music conditioned on a specific artist's style. Two evaluation protocols confirm that the generator captures meaningful stylistic structure: a sliding-window classifier consistently attributes conditioned continuations to the correct artist, far above unconditioned baselines; and a classifier trained entirely on synthetic generations identifies real pianists across 12 classes with 87% chunk-level and 95% song-level accuracy. Finally, we repurpose the classifier to locate the most characteristic moments within a performance, surfacing the specific musical gestures that distinguish each pianist's voice.
Abstract:Jazz pianists develop distinctive traits that experienced listeners can often identify within seconds, yet the features underlying this recognition resist formal description. We study jazz pianist style through the lens of a pretrained symbolic music transformer, showing that its learned representations already encode pianist identity well enough for highly accurate classification across two benchmarks. We then augment the transformer with cross-attention over learned pianist identity embeddings, enabling it to generate music conditioned on a specific artist's style. Two evaluation protocols confirm that the generator captures meaningful stylistic structure: a sliding-window classifier consistently attributes conditioned continuations to the correct artist, far above unconditioned baselines; and a classifier trained entirely on synthetic generations identifies real pianists across 12 classes with 87% chunk-level and 95% song-level accuracy. Finally, we repurpose the classifier to locate the most characteristic moments within a performance, surfacing the specific musical gestures that distinguish each pianist's voice.
8 pages, 6 figures. Accepted at ISMIR 2026. Audio demos, code and checkpoints: this https URL
Sound (cs.SD); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
arXiv:2610.02918 [cs.SD]
(or arXiv:2610.02918v1 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2610.02918 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 265 words
Related coverage
[AINews] not much happened today
Latent Space · Oct 3, 2026
Challenge: Hand coding weights for efficient sequence memorisation
AI Alignment Forum · Jul 23, 2026
NodeJEPA: Structure-Conditioned Latent Prediction for Node-Level Graph Self-Supervised Learning
arXiv · Aug 6, 2026
No Concept Escapes the Audit: Auditing-Aware Unlearning for Verifiable Concept Erasure in Diffusion Models
arXiv · Oct 7, 2026