Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
The Explainable Attention Based Deepfake Detector is a video deepfake detection tool that combines an EfficientNet CNN with a Vision Transformer architecture. It provides explainability by generating heatmap visualizations from the Transformer's attention layers, and allows users to retrain and test the model with new parameters.
Parse Score
