用概念向量发现音乐模型隐藏的性别语言偏见
Beyond Genre: Diagnosing Bias in Music Embeddings Using Concept Activation Vectors
- 用CAV方法检测音乐模型对歌手性别语言的非预期影响
- 4个主流模型均存在显著偏见，与音乐社会学研究一致
- 提出后处理去偏策略，可有效降低模型偏见
音乐表征模型广泛用于标签、检索和理解任务，但其潜在的文化偏见尚未充分研究。本文采用概念激活向量（CAVs）分析非音乐属性（如性别、语言）是否以意外方式影响音乐流派表示。在STraDa数据集上，对四个前沿模型（MERT、Whisper、MuQ、MuQ-MuLan）进行测试，通过精心平衡训练集控制流派混淆因素。结果表明，各模型存在显著的特定偏见，与音乐信息检索（MIR）和音乐社会学中的报告差异一致。进一步提出基于概念向量操作的后处理去偏策略，验证了其有效性。研究强调需构建偏见感知的模型设计，证明概念化可解释性方法是诊断和缓解音乐表征偏见的实用工具。
原文摘要 · Abstract (English)
Music representation models are widely used for tasks such as tagging, retrieval, and music understanding. Yet, their potential to encode cultural bias remains underexplored. In this paper, we apply Concept Activation Vectors (CAVs) to investigate whether non-musical singer attributes - such as gender and language - influence genre representations in unintended ways. We analyze four state-of-the-art models (MERT, Whisper, MuQ, MuQ-MuLan) using the STraDa dataset, carefully balancing training sets to control for genre confounds. Our results reveal significant model-specific biases, aligning with disparities reported in MIR and music sociology. Furthermore, we propose a post-hoc debiasing strategy using concept vector manipulation, demonstrating its effectiveness in mitigating these biases. These findings highlight the need for bias-aware model design and show that conceptualized interpretability methods offer practical tools for diagnosing and mitigating representational bias in MIR.
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。