SAGE-Music: Low-Latency Symbolic Music Generation via Attribute-Specialized Key-Value Head Sharing
Abstract
Low-latency symbolic music generation is essential for real-time improvisation and human-AI co-creation. Existing transformer-based models, however, face a trade-off between inference speed and musical quality. Traditional acceleration techniques such as embedding pooling significantly degrade quality, while recently proposed Byte Pair Encoding (BPE) methods - though effective on single-track piano data - suffer large performance drops in multi-track settings, as revealed by our analysis. We propose Attribute-Specialized Key-Value Head Sharing (AS-KVHS), adapted to music's structured symbolic representation, achieving about 30% inference speedup with only a negligible (about 0.4%) quality drop in objective evaluations and slight improvements in subjective listening tests. Our main contributions are (1) the first systematic study of BPE's generalizability in multi-track symbolic music, and (2) the introduction of AS-KVHS for low-latency symbolic music generation. Beyond these, we also release SAGE-Music, an open-source benchmark that matches or surpasses state-of-the-art models in generation quality.
Keywords
Cite
@article{arxiv.2510.00395,
title = {SAGE-Music: Low-Latency Symbolic Music Generation via Attribute-Specialized Key-Value Head Sharing},
author = {Jiaye Tan and Haonan Luo and Linfeng Song and Shuaiqi Chen and Yishan Lyu and Zian Zhong and Roujia Wang and Daniel Jiang and Haoran Zhang and Jiaming Bai and Haoran Cheng and Q. Vera Liao and Hao-Wen Dong},
journal= {arXiv preprint arXiv:2510.00395},
year = {2025}
}
Comments
Withdrawn after identifying that results in Section 5 require additional re-analysis before public dissemination