SPARK-IL reaches 94.6% mean accuracy on deepfake detection across 19 generators by fusing multi-band spectral embeddings from ViT and RGB paths, retrieving nearest signatures for majority voting, and using incremental learning with elastic weight consolidation.
C2p-clip: Injecting category common prompt in clip to enhance generalization in deepfake detection
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2roles
baseline 1polarities
baseline 1representative citing papers
Binary AI vs. real image classification reaches F1 > 0.83 while identifying the exact generative model achieves a highest F1 of 0.4986 on the MS COCOAI dataset.
citing papers explorer
-
SPARK-IL: Spectral Retrieval-Augmented RAG for Knowledge-driven Deepfake Detection via Incremental Learning
SPARK-IL reaches 94.6% mean accuracy on deepfake detection across 19 generators by fusing multi-band spectral embeddings from ViT and RGB paths, retrieving nearest signatures for majority voting, and using incremental learning with elastic weight consolidation.
-
Findings of the Counter Turing Test: AI-Generated Image Detection
Binary AI vs. real image classification reaches F1 > 0.83 while identifying the exact generative model achieves a highest F1 of 0.4986 on the MS COCOAI dataset.