首个将图像相似度用于乐谱数据复制检测的工具，可精准识别音乐生成中的抄袭。
Assessing Data Replication in Symbolic Music via Adapted Structural Similarity Index Measure
- 将乐谱转为类图像形式，用改进版结构相似性指标评估音乐复制。
- 在多数据集上验证，可精确检测至少一拍级别的完全复制。
- 适用于音乐生成模型的伦理审查，适合研究生成内容安全的研究者。
AI生成音乐可能无意中复制训练数据样本，引发抄袭担忧。相似度度量可用于量化此类复制，从而为音乐生成模型提供监督与指导。现有符号音乐相似度方法主要针对旋律重复，难以评估具有丰富织体和表现力特征的复杂音乐。为此，我们提出SSIMuse，首个将图像领域结构相似性指数（SSIM）适配至符号音乐的方法。具体地，我们将符号音乐表示为二值和速度信息的类钢琴卷轴图像；在此基础上，重新诠释并合理调整SSIM组件，构建两个变体：SSIMuse-B（用于评估创作层面复制）、SSIMuse-V（用于评估动态表现复制）。在多个数据集的合成样本上进行受控实验表明，SSIMuse可在至少一拍粒度下可靠检测完全复制。该方法支持音乐生成中的复制开放评估，并引发对更广泛伦理、社会、法律与经济影响的关注。代码已开源：https://github.com/Tayjsl97/SSIMuse。
原文摘要 · Abstract (English)
AI-generated music may inadvertently replicate samples from the training data, raising concerns of plagiarism. Similarity measures can quantify such replication, thereby offering supervision and guidance for music generation models. Existing similarity measure methods for symbolic music mainly target melody repetition, leaving a gap in assessing complex music with rich textures and expressive performance characteristics. To address this gap, we introduce SSIMuse, the first adaptation of the Structural Similarity Index Measure (SSIM) from images to symbolic music. Specifically, we represent symbolic music as image-like piano rolls in binary and velocity-based forms. Build upon these representations, we reinterprete and suitably modify the SSIM components in the musical context to develop two variants, i.e., SSIMuse-B and SSIMuse-V, for evaluating data replication in composition and dynamic performance, respectively. Controlled experiments on synthetic samples from multiple datasets show that SSIMuse can reliably detect exact replication at a granularity of at least one bar. SSIMuse enables open evaluation of replication in music generation and draws attention to its broader ethical, social, legal, and economic implications. The code is available at https://github.com/Tayjsl97/SSIMuse.
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。