Prompt-CAM
Making Vision Transformers Interpretable for Fine-Grained Analysis, published at CVPR 2025.
Research code for selecting relevant features as new dimensions arrive in a stream, with multi-label classification as the downstream task.
Making Vision Transformers Interpretable for Fine-Grained Analysis, published at CVPR 2025.
Research code for selecting relevant features as new dimensions arrive in a stream, with multi-label classification as the downstream task.