“World Knowledge” in the Weights: Reading Concept Circuits of Vision Transformers
- 1 October 2026
- book chapter
- Published by Springer Nature
- p. 169-186
Abstract
No abstract availableKeywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification2023
- A Comprehensive Study of Image Classification Model Sensitivity to Foregrounds, Backgrounds, and Visual Attributes2022
- Concept whitening for interpretable image recognitionNature Machine Intelligence, 2020
- Understanding Deep Networks via Extremal Perturbations and Smooth Masks2019
- LVIS: A Dataset for Large Vocabulary Instance Segmentation2019
- Interpretable Explanations of Black Boxes by Meaningful Perturbation2017
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization2017
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image AnnotationsInternational Journal of Computer Vision, 2017
- Learning Deep Features for Discriminative Localization2016
- ImageNet: A large-scale hierarchical image database2009