让视觉Transformer注意力更清晰可解释，还能保持高准确率。
Interpretable Vision Transformers in Image Classification via SVDA
- 用基于SVD的几何化方法改进ViT注意力机制，增强可解释性。
- 在四个数据集上验证，注意力模式更稀疏有结构，准确率不降。
- 适合关注模型解释性、注意力机制设计的研究者。
视觉变换器（ViTs）在图像分类中达到顶尖性能，但其注意力机制常不透明且呈密集无序状态。本文将先前提出的基于SVD的注意力（SVDA）机制适配至ViT架构，提出一种具有几何基础的公式，提升可解释性、稀疏性与谱结构。通过引入原始为SVDA设计的可解释性指标，监控训练过程中的注意力动态，并评估学习表征的结构性质。在四个广泛使用的基准数据集——CIFAR-10、FashionMNIST、CIFAR-100和ImageNet-100上的实验表明，SVDA始终生成更具可解释性的注意力模式，且未牺牲分类准确性。尽管当前框架仅提供描述性洞察而非指导性建议，结果确立了SVDA作为分析与开发结构化注意力模型的综合性工具。本工作为可解释AI、谱诊断及基于注意力的模型压缩奠定了基础。
原文摘要 · Abstract (English)
Vision Transformers (ViTs) have achieved state-of-the-art performance in image classification, yet their attention mechanisms often remain opaque and exhibit dense, non-structured behaviors. In this work, we adapt our previously proposed SVD-Inspired Attention (SVDA) mechanism to the ViT architecture, introducing a geometrically grounded formulation that enhances interpretability, sparsity, and spectral structure. We apply the use of interpretability indicators -- originally proposed with SVDA -- to monitor attention dynamics during training and assess structural properties of the learned representations. Experimental evaluations on four widely used benchmarks -- CIFAR-10, FashionMNIST, CIFAR-100, and ImageNet-100 -- demonstrate that SVDA consistently yields more interpretable attention patterns without sacrificing classification accuracy. While the current framework offers descriptive insights rather than prescriptive guidance, our results establish SVDA as a comprehensive and informative tool for analyzing and developing structured attention models in computer vision. This work lays the foundation for future advances in explainable AI, spectral diagnostics, and attention-based model compression.
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。