Language-Aware Visual Explanations (LAVE) is a framework designed for image classification tasks, particularly focusing on the ImageNet dataset. Unlike conventional methods that necessitate extensive training, LAVE leverages SHAP (SHapley Additive exPlanations) values to provide insightful textual and visual explanations.
deep-learning image-classification explainable-ai xai visual-explanations shap lave shapley-additive-explanations
-
Updated
Mar 26, 2026 - Jupyter Notebook