From Prediction to Explanation: Multimodal, Explainable, and Interactive Deepfake Detection Framework for Non-Expert Users

2025-08-10

AI Summary

Current deepfake detection models are predominantly black-box systems with poor interpretability, hindering trustworthy decision-making by non-expert users. To address this, we propose DF-P2E—a novel framework that integrates Grad-CAM-based saliency visualization, vision-to-language description generation, and large language model (LLM) fine-tuning for narrative optimization—thereby establishing the first multimodal, interactive explanation system tailored for lay users. DF-P2E delivers hierarchical, context-aware, and user-sensitive explanations across three levels: visual (attention maps), semantic (image-grounded captions), and narrative (coherent, natural-language reasoning). It enables seamless transition from model prediction to human-understandable inference. Evaluated on the DF40 benchmark, DF-P2E achieves detection accuracy competitive with state-of-the-art methods while generating natural language explanations tightly aligned with visual evidence. This significantly enhances model transparency, user comprehension, and human-AI collaborative decision-making capability.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

The proliferation of deepfake technologies poses urgent challenges and serious risks to digital integrity, particularly within critical sectors such as forensics, journalism, and the legal system. While existing detection systems have made significant progress in classification accuracy, they typically function as black-box models, offering limited transparency and minimal support for human reasoning. This lack of interpretability hinders their usability in real-world decision-making contexts, especially for non-expert users. In this paper, we present DF-P2E (Deepfake: Prediction to Explanation), a novel multimodal framework that integrates visual, semantic, and narrative layers of explanation to make deepfake detection interpretable and accessible. The framework consists of three modular components: (1) a deepfake classifier with Grad-CAM-based saliency visualisation, (2) a visual captioning module that generates natural language summaries of manipulated regions, and (3) a narrative refinement module that uses a fine-tuned Large Language Model (LLM) to produce context-aware, user-sensitive explanations. We instantiate and evaluate the framework on the DF40 benchmark, the most diverse deepfake dataset to date. Experiments demonstrate that our system achieves competitive detection performance while providing high-quality explanations aligned with Grad-CAM activations. By unifying prediction and explanation in a coherent, human-aligned pipeline, this work offers a scalable approach to interpretable deepfake detection, advancing the broader vision of trustworthy and transparent AI systems in adversarial media environments.

Problem

• Enhances deepfake detection interpretability for non-experts
• Combines visual, semantic, and narrative explanation layers
• Addresses black-box limitations in existing detection systems

Innovation

• Multimodal framework integrates visual, semantic, narrative layers
• Grad-CAM-based saliency visualization for deepfake classifier
• Fine-tuned LLM for context-aware, user-sensitive explanations