Defake-o3: From Speculative Rationales to Verifiable Evidence for Explainable AIGI Detection

• URL: http://arxiv.org/abs/2608.16259v1
• Date: Mon, 17 Aug 2026 08:30:32 GMT
• Title: Defake-o3: From Speculative Rationales to Verifiable Evidence for Explainable AIGI Detection
• Abstract summary: We present Defake-o3, an AIGI detector that moves from speculative rationales to verifiable evidence.<n>It combines interactive visual search with verifier-guided evidence alignment.<n>Defake-o3 improves both detection accuracy and explanation quality.
• Score: 8.599317072232198
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The rapid progress of image generation models calls for AI-generated image (AIGI) detectors that are not only accurate but also explainable and reliable. While MLLM-based detectors can provide natural language explanations, existing methods often generate speculative rationales: they rely on vague or hallucinated artifacts, miss subtle localized flaws from the latest generators, and fail to provide evidence that can be visually verified. We present Defake-o3, an explainable AIGI detector that moves from speculative rationales to verifiable evidence. It combines interactive visual search with verifier-guided evidence alignment: the model iteratively zooms into suspicious regions to inspect fine-grained details, while an Evidence Verifier, trained from human verification annotations, provides reinforcement learning rewards that favor grounded evidence and penalize baseless claims. To support this objective, we construct GroundFake, a dataset designed for grounded explainable detection, with localized bounding-box evidence, human verification based on visual grounding and artifact specificity, corrected reasoning trajectories, and valid/invalid evidence supervision. We further introduce FakeFrontier, an out-of-distribution benchmark built from real images and outputs of 10 recent generators, together with an MLLM-based protocol for evaluating evidence quality and persuasiveness. Experiments on GroundFake, FakeFrontier, and additional out-of-distribution benchmarks show that Defake-o3 improves both detection accuracy and explanation quality, producing more localized, verifiable, and persuasive evidence.

Related papers

• VidForensics-M1: Meta-Detection Reinforcement Learning with Verifiable Temporal Grounding for AI-Generated Video Forensics [21.438987128589403]
  We introduce textbfmeta-detection into AI-generated video detection, enabling reliable forgery detection.<n>Based on this insight, we propose an automated data construction pipeline that generates paired real-fake videos.<n>Experiments demonstrate that textbfVidForensics-M1 effectively leverages verifiable temporal evidence to achieve robust and generalizable AI-generated video detection.
  arXiv  Detail & Related papers  (2026-08-11T17:58:10Z)
• Grounding and Explaining Visual Evidence for AI-Generated Image Detection in Human-Centric Scenes [33.05351037039477]
  We present HAVE (Human-centric AI-generated Visual Evidence), a diverse human-centric dataset comprising 40K real and 39K AI-generated images.<n>We propose PAVE, a Perception-Aware Visual Evidence framework that jointly performs authenticity prediction, visual evidence grounding, and explanation generation.<n>Experiments on HAVE and external datasets demonstrate strong performance in authenticity detection, visual evidence grounding, and explanation quality.
  arXiv  Detail & Related papers  (2026-08-03T09:49:24Z)
• XPlainVerse: A Million-Scale Benchmark for Explainable Deepfake Detection [43.15397857869723]
  We introduce XPlainVerse, a large-scale benchmark for joint deepfake detection and human-centered explanation.<n>XPlainVerse comprises one million real and manipulated images, pairing authentic images from five established sources with forgeries generated by twelve off-the-shelf image editing and synthesis models.<n>Beyond dataset scale, XPlainVerse provides two complementary explanation styles: technical explanations for expert analysis and simplified explanations optimized for non-technical users.
  arXiv  Detail & Related papers  (2026-07-03T18:52:42Z)
• FraudBench: A Multimodal Benchmark for Detecting AI-Generated Fraudulent Refund Evidence [36.982811727399984]
  FraudBench is a multimodal benchmark for detecting AI-generated fraudulent refund evidence.<n>It is constructed from real-world user-review evidence across e-commerce, food delivery, and travel-service scenarios.<n>We evaluate MLLMs, specialized AI-generated image detectors, and human participants under the same settings.
  arXiv  Detail & Related papers  (2026-05-09T09:11:15Z)
• Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detection [95.08316274158165]
  X-AIGD provides pixel-level, categorized annotations of perceptual artifacts, spanning low-level distortions, high-level semantics, and cognitive-level counterfactuals.<n>Existing AIGI detectors demonstrate negligible reliance on perceptual artifacts, even at the most basic distortion level.<n>Explicitly aligning model attention with artifact regions can increase the interpretability and generalization of detectors.
  arXiv  Detail & Related papers  (2026-01-27T10:09:17Z)
• REVEAL: Reasoning-enhanced Forensic Evidence Analysis for Explainable AI-generated Image Detection [30.963994372913092]
  We introduce textbfREVEAL-Bench, the first reasoning-enhanced multimodal benchmark for AI-generated image detection.<n>Our framework integrates detection with a novel expert-grounded reinforcement learning.<n> REVEAL significantly enhances detection accuracy, explanation fidelity, and robust cross-model generalization.
  arXiv  Detail & Related papers  (2025-11-28T13:11:08Z)
• Look As You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning [55.232400251303794]
  Look As You Think (LAT) is a reinforcement learning framework that trains models to produce verifiable reasoning paths with consistent attribution.<n>LAT consistently improves the vanilla model in both single- and multi-image settings, yielding average gains of 8.23% in soft exact match (EM) and 47.0% in IoU@0.5.
  arXiv  Detail & Related papers  (2025-11-15T02:50:23Z)
• Semantic Visual Anomaly Detection and Reasoning in AI-Generated Images [96.43608872116347]
  AnomReason is a large-scale benchmark with structured annotations as quadruple textbfAnomAgent<n>AnomReason and AnomAgent serve as a foundation for measuring and improving the semantic plausibility of AI-generated images.
  arXiv  Detail & Related papers  (2025-10-11T14:09:24Z)
• Seeing Before Reasoning: A Unified Framework for Generalizable and Explainable Fake Image Detection [58.82268659497348]
  We argue that the root of this failure lies in a fundamental mismatch: MLLMs are asked to reason about fakes before they can truly see them.<n>We propose Forensic-Chat, a generalizable, explainable, and still-conversational assistant for fake image detection.
  arXiv  Detail & Related papers  (2025-09-29T20:59:19Z)
• Evaluating Evidence Attribution in Generated Fact Checking Explanations [48.776087871960584]
  We introduce a novel evaluation protocol -- citation masking and recovery.<n>We implement our protocol using both human annotators and automatic annotators.<n>Experiments reveal that the best-performing LLMs still generate explanations with inaccurate attributions.
  arXiv  Detail & Related papers  (2024-06-18T14:13:13Z)
• Read it Twice: Towards Faithfully Interpretable Fact Verification by Revisiting Evidence [59.81749318292707]
  We propose a fact verification model named ReRead to retrieve evidence and verify claim. The proposed system is able to achieve significant improvements upon best-reported models under different settings.
  arXiv  Detail & Related papers  (2023-05-02T03:23:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.