MI2DataLab
diff --git a/‎2025/2025_05_05_diffex/Explaining in Diffusion- Explaining a Classifier Through Hierarchical Semantics with Text-to-Image Diffusion Models.pdf‎
3.68 MB b/‎2025/2025_05_05_diffex/Explaining in Diffusion- Explaining a Classifier Through Hierarchical Semantics with Text-to-Image Diffusion Models.pdf‎
3.68 MB
diff --git a/‎2025/2025_05_05_diffex/README.md‎
Lines changed: 3 additions & 0 deletions b/‎2025/2025_05_05_diffex/README.md‎
Lines changed: 3 additions & 0 deletions
@@ -0,0 +1,3 @@
+# Explaining in Diffusion - Explaining a Classifier Through Hierarchical Semantics with Text-to-Image Diffusion Models
+
+Presentation introduces DiffEx, a method for explaining image classifier decisions using text-to-image diffusion models. By leveraging vision-language models, DiffEx builds a hierarchical semantic structure to identify and edit influential visual features without retraining. The approach is domain-agnostic, supports complex scenes, and enables both individual and joint attribute analysis.
Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	`+# Explaining in Diffusion - Explaining a Classifier Through Hierarchical Semantics with Text-to-Image Diffusion Models`
	`2`	`+`
	`3`	`+Presentation introduces DiffEx, a method for explaining image classifier decisions using text-to-image diffusion models. By leveraging vision-language models, DiffEx builds a hierarchical semantic structure to identify and edit influential visual features without retraining. The approach is domain-agnostic, supports complex scenes, and enables both individual and joint attribute analysis.`