LAA-X: Unified Localized Artifact Attention for Quality-Agnostic and Generalizable Face Forgery Detection

TL;DR

LAA-X is a new deepfake detection framework that uses explicit localized artifact attention and multi-task learning to improve robustness and generalization to unseen manipulations.

Contribution

It introduces an explicit attention strategy with multi-task learning and blending-based data synthesis, compatible with CNN and transformer backbones, enhancing deepfake detection.

Findings

LAA-X performs competitively with state-of-the-art methods across benchmarks.

The framework is effective even when trained only on real and pseudo-fake samples.

Abstract

In this paper, we propose Localized Artifact Attention X (LAA-X), a novel deepfake detection framework that is both robust to high-quality forgeries and capable of generalizing to unseen manipulations. Existing approaches typically rely on binary classifiers coupled with implicit attention mechanisms, which often fail to generalize beyond known manipulations. In contrast, LAA-X introduces an explicit attention strategy based on a multi-task learning framework combined with blending-based data synthesis. Auxiliary tasks are designed to guide the model toward localized, artifact-prone (i.e., vulnerable) regions. The proposed framework is compatible with both CNN and transformer backbones, resulting in two different versions, namely, LAA-Net and LAA-Former, respectively. Despite being trained only on real and pseudo-fake samples, LAA-X competes with state-of-the-art methods across multiple benchmarks. Code and pre-trained weights for LAA-Net\footnote{https://github.com/10Ring/LAA-Net} and LAA-Former\footnote{https://github.com/10Ring/LAA-Former} are publicly available.