# Pixels Don't Lie (But Your Detector Might): Bootstrapping MLLM-as-a-Judge for Trustworthy Deepfake Detection and Reasoning Supervision

arXiv: https://arxiv.org/abs/2602.19715 · HTML: https://mlsanity.com/paper/2602.19715/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 53 references that are in the map.
- First posted on arXiv 23 Feb 2026.

## Builds on (most cited first, 30 of 31)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models](https://mlsanity.com/paper/2306.13394/) (2023)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [FaceForensics++: Learning to Detect Manipulated Facial Images](https://mlsanity.com/paper/1901.08971/) (2019)
- [Face X-ray for More General Face Forgery Detection](https://mlsanity.com/paper/1912.13458/) (2019)
- [Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuning](https://mlsanity.com/paper/2306.14565/) (2023)
- [DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models](https://mlsanity.com/paper/2210.14896/) (2022)
- [Large Language Models are Better Reasoners with Self-Verification](https://mlsanity.com/paper/2212.09561/) (2022)
- [Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributions](https://mlsanity.com/paper/2003.01826/) (2020)
- [DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection](https://mlsanity.com/paper/2001.03024/) (2020)
- [Universal Self-Consistency for Large Language Model Generation](https://mlsanity.com/paper/2311.17311/) (2023)
- [Capsule-Forensics: Using Capsule Networks to Detect Forged Images and Videos](https://mlsanity.com/paper/1810.11215/) (2018)
- [Transcending Forgery Specificity with Latent Space Augmentation for Generalizable Deepfake Detection](https://mlsanity.com/paper/2311.11278/) (2023)
- [FakeShield: Explainable Image Forgery Detection and Localization via Multi-modal Large Language Models](https://mlsanity.com/paper/2410.02761/) (2024)
- [SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model](https://mlsanity.com/paper/2412.04292/) (2024)
- [SEED-Bench-2: Benchmarking Multimodal Large Language Models](https://mlsanity.com/paper/2311.17092/) (2023)
- [HalluLens: LLM Hallucination Benchmark](https://mlsanity.com/paper/2504.17550/) (2025)
- [Exploiting Style Latent Flows for Generalizing Deepfake Video Detection](https://mlsanity.com/paper/2403.06592/) (2024)
- [Reducing Hallucinations in Vision-Language Models via Latent Space Steering](https://mlsanity.com/paper/2410.15778/) (2024)
- [Community Forensics: Using Thousands of Generators to Train Fake Image Detectors](https://mlsanity.com/paper/2411.04125/) (2024)
- [Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference](https://mlsanity.com/paper/2211.11875/) (2022)
- [BusterX++: Towards Unified Cross-Modal AI-Generated Content Detection and Explanation with MLLM](https://mlsanity.com/paper/2507.14632/) (2025)
- [Towards Relatable Explainable AI with the Perceptual Process](https://mlsanity.com/paper/2112.14005/) (2021)

## Built on by (most cited first, 5 of 5)

- [XPlainVerse: A Million-Scale Benchmark for Explainable Deepfake Detection](https://mlsanity.com/paper/2607.03562/) (2026)
- [SPARED: Reasoning-Based AI-Generated Image Detection via Adversarially Edited Data](https://mlsanity.com/paper/2608.12876/) (2026)
- [Explainable Deepfake Detection with Feature-robust Augmentation and Evidence-grounded Explanation Optimization](https://mlsanity.com/paper/2608.20913/) (2026)
- [Beyond Semantics: Uncovering the Physics of Fakes via Universal Physical Descriptors for Cross-Modal Synthetic Detection](https://mlsanity.com/paper/2604.04608/) (2026)
- [Mandela-Bench: Multimodal Models Remember Canonical Images Instead of Seeing Them](https://mlsanity.com/paper/2609.32763/) (2026)

## Abstract (arXiv)

Deepfake detection models often generate natural-language explanations, yet their reasoning is frequently ungrounded in visual evidence, limiting reliability. Existing evaluations measure classification accuracy but overlook reasoning fidelity. We propose DeepfakeJudge, a framework for scalable reasoning supervision and evaluation, that integrates an out-of-distribution benchmark containing recent generative and editing forgeries, a human-annotated subset with visual reasoning labels, and a suite of evaluation models, that specialize in evaluating reasoning rationales without the need for explicit ground truth reasoning rationales. The Judge is optimized through a bootstrapped generator-evaluator process that scales human feedback into structured reasoning supervision and supports both pointwise and pairwise evaluation. On the proposed meta-evaluation benchmark, our reasoning-bootstrapped model achieves an accuracy of 96.2%, outperforming texttt30x larger baselines. The reasoning judge attains very high correlation with human ratings and 98.9% percent pairwise agreement on the human-annotated meta-evaluation subset. These results establish reasoning fidelity as a quantifiable dimension of deepfake detection and demonstrate scalable supervision for interpretable deepfake reasoning. Our user study shows that participants preferred the reasonings generated by our framework 70% of the time, in terms of faithfulness, groundedness, and usefulness, compared to those produced by other models and datasets. All of our datasets, models, and codebase are hrefhttps://github.com/KjAeRsTuIsK/DeepfakeJudgeopen-sourced.