20.500.12556/RUL-183410 Learning the manifold of authenticity hybrid-curvature representation learning for generalizable deepfake detection The practical utility of deepfake detectors is crippled by a crisis of generalization: models that perform well on known manipulation techniques consistently fail when faced with unseen forgeries.We argue this failure stems from a fundamental geometric mismatch. Existing methods implicitly assume that the manifold of authentic faces can be modeled in a space of uniform curvature, typically Euclidean, which inade-quately captures the complex, multi-scale structure of facial features. This paper validates the hypothesis that authentic faces lie on a manifold whose geometry is inherently hybrid, requiring both angular compactness (a spherical property) and hierarchical organization (a hyperbolic property). To resolve this geometric mismatch, we introduce a novel detector, CTrue, that learns a unified, hybrid-curvature representation of facial authenticity. Trained exclusively on real faces via self-supervised learning, our method simultaneously projects facial embeddings onto two complementary manifolds: a hypersphere to enforce compactness and a hyperbolic space to model the natural feature hierarchy. A single set of mathematically-optimal prototypes acts as a ‘‘geometric bridge’’, unifying the learning objectives in both spaces. At inference, a composite score measures an embedding’s deviation from this learned manifold. On challenging cross-dataset and cross-manipulation benchmarks, our method achieves competitive generalization under a strictly pristine-only training setting, showing that hybrid-curvature representations provide an effective and data-efficient alternative for deepfake detection. contrast learning generalized deepfake detection hyperbolic one-class anomaly detection kontrastno učenje posplošena detekcija globokih ponaredkov hiperboličnost eno-razredna detekcija anomalij true false true Angleški jezik Slovenski jezik Članek v reviji 2026-06-12 11:45:21 2026-06-12 11:45:22 2026-06-13 04:29:47 0000-00-00 00:00:00 2026 0 0 Str. 1-14 Vol. 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.932.2:004.8 2169-3536 10.1109/ACCESS.2026.3702429 281381123 RAZ_Larue_Nicolas_2026.pdf RAZ_Larue_Nicolas_2026.pdf 1 5729AAA4AF4B867AC73D8368460B548C 6fb69d02c8b4717620300aed9cb3bf466dc7f9086a2fd5800238ada4ef41c5fe f28b654e-6643-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=235456 https://ieeexplore.ieee.org/document/11557307 1 677676aa-6643-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=235455 Fakulteta za računalništvo in informatiko Fakulteta za elektrotehniko 0 0 0