语音分类中主流特征归因方法大多不可靠，仅词级扰动有效。
On the reliability of feature attribution methods for speech classification
- 对比输入类型、扰动时长与聚合方式对归因可靠性的影响
- 多数方法在语音任务中不可靠，仅词对齐扰动在词分类任务有效
- 研究结果对模型可解释性评估有重要指导意义
随着大规模预训练模型能力的提升，理解其输出决定因素变得愈发重要。特征归因旨在揭示输入中哪些部分对模型输出贡献最大。在语音处理中，输入信号的独特特性使得特征归因方法的应用面临挑战。本文研究了输入类型、聚合方式及扰动时长等因素如何影响标准特征归因方法的可靠性，并探讨这些因素与各类分类任务特性的交互作用。研究发现，将标准归因方法应用于语音领域时普遍不可靠，仅在词级分类任务中使用词对齐扰动方法时表现可靠。
原文摘要 · Abstract (English)
As the capabilities of large-scale pre-trained models evolve, understanding the determinants of their outputs becomes more important. Feature attribution aims to reveal which parts of the input elements contribute the most to model outputs. In speech processing, the unique characteristics of the input signal make the application of feature attribution methods challenging. We study how factors such as input type and aggregation and perturbation timespan impact the reliability of standard feature attribution methods, and how these factors interact with characteristics of each classification task. We find that standard approaches to feature attribution are generally unreliable when applied to the speech domain, with the exception of word-aligned perturbation methods when applied to word-based classification tasks.
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。