为音频分类模型的时序解释构建评估基准，提升可解释性。
Benchmarking Time-localized Explanations for Audio Classification Models
- 用目标事件的时间标注作为解释真值，构建时序解释评估基准。
- 部分方法优化后可生成接近完美的时序解释结果。
- 可用于发现模型中的虚假关联，提升模型可信度。
当前多数音频处理方法缺乏透明度，无法提供决策依据。为此，研究者提出了多种解释方法以揭示模型输出。优质解释能带来数据或模型的新洞察，并增强系统可信度。然而，由于多数任务缺乏明确的真值解释，解释质量评估极为困难。本文提出一个面向音频分类模型的时序局部解释基准，以目标事件的时间标注作为近似真值。利用该基准，系统性地优化与比较多种模型无关的后验解释方法，部分情况下获得近乎完美的解释结果。最后，展示了这些解释在揭示虚假相关性方面的实际价值。
原文摘要 · Abstract (English)
Most modern approaches for audio processing are opaque, in the sense that they do not provide an explanation for their decisions. For this reason, various methods have been proposed to explain the outputs generated by these models. Good explanations can result in interesting insights about the data or the model, as well as increase trust in the system. Unfortunately, evaluating the quality of explanations is far from trivial since, for most tasks, there is no clear ground truth explanation to use as reference. In this work, we propose a benchmark for time-localized explanations for audio classification models that uses time annotations of target events as a proxy for ground truth explanations. We use this benchmark to systematically optimize and compare various approaches for model-agnostic post-hoc explanation, obtaining, in some cases, close to perfect explanations. Finally, we illustrate the utility of the explanations for uncovering spurious correlations.
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。