Picture for Sangyu Han

Sangyu Han

Bi-ICE: An Inner Interpretable Framework for Image Classification via Bi-directional Interactions between Concept and Input Embeddings

Add code
Nov 26, 2024
Viaarxiv icon

Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)

Add code
Sep 03, 2024
Figure 1 for Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)
Figure 2 for Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)
Figure 3 for Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)
Figure 4 for Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)
Viaarxiv icon

Respect the model: Fine-grained and Robust Explanation with Sharing Ratio Decomposition

Add code
Jan 25, 2024
Viaarxiv icon