REVEAL: Reasoning-enhanced Forensic Evidence Analysis for Explainable AI-generated Image Detection

2025-11-28

AI Summary

To address the declining interpretability of digital forensics due to increasingly realistic AI-generated images, this paper proposes the first multimodal explainable detection framework grounded in verifiable evidence chains. Methodologically, we introduce REVEAL-Bench—a novel benchmark dataset—and pioneer an expert-model-driven reinforcement learning mechanism that jointly integrates lightweight expert models, multimodal reasoning tracing, and causality-guided policies to co-generate detection decisions and human-interpretable explanations. Our key contributions are threefold: (1) the first multimodal evidence-chain benchmark for AI-image forensics; (2) a causally consistent and logically coherent explainability paradigm; and (3) significant improvements in detection accuracy, explanation fidelity, and cross-model generalization—outperforming state-of-the-art explainable forensic methods across multiple quantitative metrics.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

With the rapid advancement of generative models, visually realistic AI-generated images have become increasingly difficult to distinguish from authentic ones, posing severe threats to social trust and information integrity. Consequently, there is an urgent need for efficient and truly explainable image forensic methods. Recent detection paradigms have shifted towards explainable forensics. However, state-of-the-art approaches primarily rely on post-hoc rationalizations or visual discrimination, lacking a verifiable chain of evidence. This reliance on surface-level pattern matching limits the generation of causally grounded explanations and often results in poor generalization. To bridge this critical gap, we introduce extbf{REVEAL-Bench}, the first reasoning-enhanced multimodal benchmark for AI-generated image detection that is explicitly structured around a chain-of-evidence derived from multiple lightweight expert models, then records step-by-step reasoning traces and evidential justifications. Building upon this dataset, we propose extbf{REVEAL} (underline{R}easoning-underline{e}nhanced Forensic Eunderline{v}idunderline{e}nce underline{A}naunderline{l}ysis), an effective and explainable forensic framework that integrates detection with a novel expert-grounded reinforcement learning. Our reward mechanism is specially tailored to jointly optimize detection accuracy, explanation fidelity, and logical coherence grounded in explicit forensic evidence, enabling REVEAL to produce fine-grained, interpretable, and verifiable reasoning chains alongside its detection outcomes. Extensive experimental results demonstrate that REVEAL significantly enhances detection accuracy, explanation fidelity, and robust cross-model generalization, benchmarking a new state of the art for explainable image forensics.

Problem

• Detecting AI-generated images that are visually indistinguishable from authentic ones
• Addressing the lack of verifiable evidence chains in current detection methods
• Improving generalization beyond surface-level pattern matching in forensic analysis

Innovation

• Uses reasoning-enhanced multimodal benchmark for detection
• Integrates expert-grounded reinforcement learning framework
• Optimizes accuracy and explanation fidelity with evidence