Search papers, labs, and topics across Lattice.
This study explores the limitations of current deepfake detection methods, which often fail to provide meaningful explanations for their predictions. By employing Encoding-Decoding Direction Pairs (EDDP), the authors uncover the semantic vocabulary of deepfake detectors, revealing the implicit features that distinguish real from fake content. The findings enhance the interpretability of detection models, enabling better understanding and application in critical contexts like legal proceedings.
Uncovering the hidden semantic vocabulary of deepfake detectors transforms our understanding of how these models differentiate between real and fake content.
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.
