Loading
Life sciences · Preprint
arXiv · September 4, 2026
Early or partial results. Treat as a signal, not a conclusion.
AdaGate-DF is a novel adaptive gated architecture for deepfake detection that routes images by quality to reduce computational cost. Performance is reported as AUC on two benchmark datasets, with higher AUC at higher resolution. This is an unpublished technical study with no independent validation, clinical application, or evidence of real-world utility.
Comparative algorithm validation on benchmark datasets. Synthetic face images from two benchmark datasets: Celeb-DF and FaceForensics++.. Intervention: AdaGate-DF: adaptive gated deepfake detection framework using image-quality cues and dual multi-exit routing.. Compared with: MaD-CoRN, DefakeHop++, and ShuffleNetV2 deepfake detection models..
On Celeb-DF, AdaGate-DF achieves AUC of 0.9370, outperforming MaD-CoRN and DefakeHop++ At 384 × 384 resolution, AUC reaches 0.9708 on Celeb-DF FaceForensics++ results show competitive performance under class imbalance
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
No clinical application is described or applicable. This is a computer vision algorithm validation study; it does not address detection or prevention of deepfake-related clinical harms.
Single-center technical validation of a novel machine learning architecture on benchmark datasets with surrogate endpoints (AUC); no clinical application, no peer review, and no independent external validation reported.
As stated by the source record.
Quoted from the source exactly as published.
No clinical application is described or applicable. This is a computer vision algorithm validation study; it does not address detection or prevention of deepfake-related clinical harms.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Deepfake detection models often rely on high-quality inputs, fixed inference paths, and computationally expensive architectures, limiting their use in low-resolution and resource-constrained settings. This paper proposes AdaGate-DF, an adaptive gated deepfake detection framework that uses image-quality cues to route samples through a dual multi-exit system so high-quality images can exit earlier and save compute. We evaluated AdaGate-DF against MaD-CoRN, DefakeHop++, and ShuffleNetV2 on two benchmark datasets (Celeb-DF and FaceForensics++) under multiple configurations to test image resolution dependence and training and inference efficiency. On Celeb-DF, AdaGate-DF achieves an AUC of 0.9370, outperforming MaD-CoRN and DefakeHop++ while maintaining a low inference latency. Resolution-based testing shows consistent improvement as input resolution increases, reaching an AUC of 0.9708 at 384 by 384. The FaceForensics++ results highlight that AdaGate-DF remains effective under class imbalance, following competitive results with evaluated models. Overall, AdaGate-DF demonstrated a practical balance between detection performance, uncertainty-aware prediction, and computational efficiency for variable-quality deepfake detection.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.