Why Fake ? Unveiling the Semantic Vocabulary of Deepfake Detectors
For researchers and practitioners in deepfake detection, this work provides a novel explainability approach that addresses the lack of interpretability in black-box detectors, but it is incremental as it applies an existing XAI technique to a new domain.
The paper introduces a method using Encoding-Decoding Direction Pairs (EDDP) to uncover the semantic vocabulary of deepfake detectors, revealing hidden real and fake features learned during training and enabling nuanced explanations without human annotations.
Deepfake (DF) technology poses a significant threat to information integrity, driving the need for robust detection methods. Most DF detectors only consider predicting a binary label for whether the input is real or fake, lacking the justification required for real-world applications like legal proceedings. Explainable DF Detection has emerged to address this limitation, but existing techniques frequently fall short by either relying on human annotations for precise artifact localization or generating superficially plausible textual explanations without grounding. This work investigates the use of post-hoc explainable AI (XAI) to analyze the decision-making process of state-of-the-art black-box DF detectors. Specifically, we employ Encoding-Decoding Direction Pairs (EDDP), a technique suitable for uncovering the concept space of DF detectors (their semantic vocabulary) as well as the mechanism for writing and reading concept information to and from internal representations. Our analysis reveals previously hidden real and fake features learned implicitly during detector training, offering nuanced explanations unattainable through conventional methods. This enables global model understanding, spatially aware concept localization, and counterfactual what-if analysis, all contributing to a deeper comprehension of DF detection strategies.