ViT-Explainer: An Interactive Walkthrough of the Vision Transformer Pipeline
This addresses the problem of interpretability for researchers and practitioners using Vision Transformers, but it is incremental as it builds on existing visualization tools.
The paper tackled the challenge of understanding Vision Transformer inference by developing ViT-Explainer, an interactive web-based system that visualizes the full pipeline from patch tokenization to classification, with a user study showing it helps users interpret model behavior.
Transformer-based architectures have become the shared backbone of natural language processing and computer vision. However, understanding how these models operate remains challenging, particularly in vision settings, where images are processed as sequences of patch tokens. Existing interpretability tools often focus on isolated components or expert-oriented analysis, leaving a gap in guided, end-to-end understanding of the full inference pipeline. To bridge this gap, we present ViT-Explainer, a web-based interactive system that provides an integrated visualization of Vision Transformer inference, from patch tokenization to final classification. The system combines animated walkthroughs, patch-level attention overlays, and a vision-adapted Logit Lens within both guided and free exploration modes. A user study with six participants suggests that ViT-Explainer is easy to learn and use, helping users interpret and understand Vision Transformer behavior.