Machine learningDeep learning / NLP / CV
可解释视觉 Transformer
可解释视觉 Transformer (Explainable Vision Transformer, XVT) 将视觉 Transformer (ViT) 强大的图像识别性能与归因技术（如相关性传播、注意力展开或梯度加权注意力）相结合，以突出显示驱动每次预测的图像区域。该方法使研究人员和从业人员能够在不牺牲准确性的前提下审计模型决策并满足透明度要求。
阅读完整方法
仅限会员
登录使用免费账户登录即可阅读本节。
Method map
The neighbourhood of related methods — select a node to explore.
来源
- Chefer, H., Gur, S., & Wolf, L. (2021). Transformer interpretability beyond attention visualization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 782–791. DOI: 10.1109/CVPR46437.2021.00084 ↗
- Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., … Houlsby, N. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. In International Conference on Learning Representations (ICLR). link ↗
如何引用本页
ScholarGate. (2026, June 3). Explainable Vision Transformer (XViT / ViT with Post-hoc Attribution). ScholarGate. https://scholargate.app/zh/deep-learning/explainable-vision-transformer
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- 图像分类深度学习↔ compare
- 多模态视觉变换器深度学习↔ compare
- 自监督视觉Transformer深度学习↔ compare
- 语义分割深度学习↔ compare
- Vision Transformer深度学习↔ compare