← back to paper
arxiv: 2605.01790 · 2 revisions
Shao: Scaling Acoustic Token Language Models Toward High-Fidelity Music Generation