# Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles Source: [Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles]() Collection index: [llms.txt]() Data updated: 2026-08-04T05:00:05.533Z Authors: Minh-Khoa Le-Phan, Minh-Hoang Le, Trong-Le Do, Minh-Triet Tran Published: 2026-04-28T17:32:48.000Z ## Abstract Current deepfake detection models achieve state-of-the-art performance on pristine academic datasets but suffer severe spatial attention drift under real-world compound degradations, such as blurring and severe lossy compression. To address this vulnerability, we propose a foundation-driven forensic framework that integrates an extreme compound degradation engine with a structurally constrained, multi-stream architecture. During training, our degradation pipeline systematically destroys high-frequency artifacts, optimizing the DINOv2-Giant backbone to extract invariant geometric and semantic priors. We then process images through three specialized pathways: a Global Texture stream, a Localized Facial stream, and a Hybrid Semantic Fusion stream incorporating CLIP. Through analyzing spatial attribution via Score-CAM and feature stability using Cosine Similarity, we quantitatively demonstrate that these streams extract non-redundant, complementary feature representations and stabilize attention entropy. By aggregating these predictions via a calibrated, discretized voting mechanism, our ensemble successfully suppresses background attention drift while acting as a robust geometric anchor. Our approach yields highly stable zero-shot generalization, achieving Fourth Place in the NTIRE 2026 Robust Deepfake Detection Challenge at CVPR. Code is available at https://github.com/khoalephanminh/ntire26-deepfake-challenge. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Deepfake Detection on NTIRE Deepfake Detection Challenge Public 2026 (test)](): Deepfake Detection; NTIRE Deepfake Detection Challenge Public 2026 (test); AUC: 87.75. - [Deepfake Detection on NTIRE Deepfake Detection Challenge Private 2026 (test)](): Deepfake Detection; NTIRE Deepfake Detection Challenge Private 2026 (test); AUC: 0.8523. ## Original sources and implementation [Original paper]() [PDF]()