音乐基础模型用螺旋结构编码音高，揭示其内在几何机制。
Do Music Foundation Models Embed Pitch in Helical Structure?
- 通过分析模型中间层表示，发现音高以螺旋结构呈现。
- 螺旋结构的清晰度随模型和输入声学特性变化。
- 为理解音乐模型内部机制提供新视角，适合研究者参考。
本研究分析了音乐基础模型（MFMs）的中间表示，揭示了其用于表征音高信息的几何结构。通过向训练好的MFMs输入孤立音符并分析其主成分，我们发现这些表示形成了反映音高八度周期性的螺旋结构。此外，我们表明该螺旋结构的清晰度和几何特征不仅在不同模型间存在差异，还随输入信号的声学特性而变化。本研究为阐明MFMs的内部工作机制提供了新方法。
原文摘要 · Abstract (English)
This study analyzes the intermediate representations of music foundation models (MFMs) and reports the geometric structures used to represent pitch information. By inputting isolated musical notes into trained MFMs and analyzing their principal components, we reveal that the representations form a helical structure reflecting the octave periodicity of pitch. Furthermore, we show that the clarity and geometry of this helical structure vary not only across models but also with the acoustic properties of the input signals. Our analysis provides a novel approach for clarifying the internal mechanisms of MFMs.
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。