通过时间错位调度提升音乐生成效率与质量平衡
Time-Shifted Token Scheduling for Symbolic Music Generation
- 用时间错位策略分步展开复合音符，建模内部依赖
- 在紧凑表示下提升所有指标，逼近细粒度表示效果
- 无需额外参数，可无缝集成到现有模型中
符号化音乐生成面临效率与质量的根本权衡。细粒度标记法虽能保证强连贯性，但导致序列过长、计算复杂；紧凑标记法虽提升效率，却牺牲了词内依赖关系。为此，我们引入基于延迟的调度机制（DP），将复合类标记跨解码步骤逐步展开，实现对词内依赖的自回归建模，同时保持高效性。该方法为轻量级策略，不引入额外参数，可无缝集成至现有表示框架。在符号化管弦乐MIDI数据集上的实验表明，该方法在所有指标上均优于标准复合标记法，并缩小了与细粒度标记法之间的差距。
原文摘要 · Abstract (English)
Symbolic music generation faces a fundamental trade-off between efficiency and quality. Fine-grained tokenizations achieve strong coherence but incur long sequences and high complexity, while compact tokenizations improve efficiency at the expense of intra-token dependencies. To address this, we adapt a delay-based scheduling mechanism (DP) that expands compound-like tokens across decoding steps, enabling autoregressive modeling of intra-token dependencies while preserving efficiency. Notably, DP is a lightweight strategy that introduces no additional parameters and can be seamlessly integrated into existing representations. Experiments on symbolic orchestral MIDI datasets show that our method improves all metrics over standard compound tokenizations and narrows the gap to fine-grained tokenizations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。