mlsanityData to 14 Oct 2026

FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Models

arXiv 2404.13306 ↗First posted 20 Apr 2024cs.CV cs.MMcited by 36

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

Abstract (arXiv)

The ability to distinguish whether an image is generated by artificial intelligence (AI) is a crucial ingredient in human intelligence, usually accompanied by a complex and dialectical forensic and reasoning process. However, current fake image detection models and databases focus on binary classification without understandable explanations for the general populace. This weakens the credibility of authenticity judgment and may conceal potential model biases. Meanwhile, large multimodal models (LMMs) have exhibited immense visual-text capabilities on various tasks, bringing the potential for explainable fake image detection. Therefore, we pioneer the probe of LMMs for explainable fake image detection by presenting a multimodal database encompassing textual authenticity descriptions, the FakeBench. For construction, we first introduce a fine-grained taxonomy of generative visual forgery concerning human perception, based on which we collect forgery descriptions in human natural language with a human-in-the-loop strategy. FakeBench examines LMMs with four evaluation criteria: detection, reasoning, interpretation and fine-grained forgery analysis, to obtain deeper insights into image authenticity-relevant capabilities. Experiments on various LMMs confirm their merits and demerits in different aspects of fake image detection tasks. This research presents a paradigm shift towards transparency for the fake image detection area and reveals the need for greater emphasis on forensic elements in visual-language research and AI risk control. FakeBench will be available at https://github.com/Yixuan423/FakeBench.

Key facts. Cited by 36 papers in the map (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 94 of its 133 references that are in the map. First posted on arXiv 20 Apr 2024.

Builds on

94 of 133 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics21 Mar
  2. 2023AntifakePrompt: Prompt-Tuned Vision-Language Models are Fake Image Detectors26 Oct

Other references · 92

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  6. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  7. 2023Visual Instruction Tuningcited by 7,869
  8. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  9. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  10. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  11. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  12. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
38 more
  1. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  2. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  3. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,919
  4. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  5. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  6. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,355
  7. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,239
  8. 2023Kosmos-2: Grounding Multimodal Large Language Models to the Worldcited by 1,089
  9. 2021GLM: General Language Model Pretraining with Autoregressive Blank Infillingcited by 1,031
  10. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  11. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 890
  12. 2019CNN-generated images are surprisingly easy to spot... for nowcited by 847
  13. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 774
  14. 2023Multimodal Chain-of-Thought Reasoning in Language Modelscited by 752
  15. 2023Can Large Language Models Be an Alternative to Human Evaluations?cited by 734
  16. 2016The Mythos of Model Interpretabilitycited by 721
  17. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 688
  18. 2021Learning To Retrieve Prompts for In-Context Learningcited by 648
  19. 2023Otter: A Multi-Modal Model with In-Context Instruction Tuningcited by 550
  20. 2023mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaborationcited by 547
  21. 2020Leveraging Frequency Analysis for Deep Fake Image Recognitioncited by 504
  22. 2023Towards Universal Fake Image Detectors that Generalize Across Generative Modelscited by 490
  23. 2023DIRE for Diffusion-Generated Image Detectioncited by 390
  24. 2022DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Modelscited by 379
  25. 2023DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Datacited by 337
  26. 2023Explainability for Large Language Models: A Surveycited by 333
  27. 2024InternLM-XComposer2: Mastering Free-form Text-Image Composition and Comprehension in Vision-Language Large Modelcited by 326
  28. 2023GenImage: A Million-Scale Benchmark for Detecting AI-Generated Imagecited by 321
  29. 2023OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documentscited by 316
  30. 2022CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformerscited by 307
  31. 2022On the detection of synthetic images generated by diffusion modelscited by 283
  32. 2020What makes fake images detectable? Understanding properties that generalizecited by 273
  33. 2023Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Visioncited by 249
  34. 2020Global Texture Enhancement for Fake Face Detection in the Wildcited by 220
  35. 2021A Survey of Human-in-the-loop for Machine Learningcited by 215
  36. 2023OpenAGI: When LLM Meets Domain Expertscited by 197
  37. 2022DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Modelscited by 168
  38. 2023Raising the Bar of AI-generated Image Detection with CLIPcited by 164

50 most cited of 92 All 92 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 10; 2026: 20.

36 papers.

Sort:

By area:

  1. 2024MME-Survey: A Comprehensive Survey on Evaluation of Multimodal LLMscited by 72
  2. 2024A Bias-Free Training Paradigm for More General AI-generated Image Detectioncited by 67
  3. 2024LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Modelscited by 51
  4. 2024X2-DFD: A framework for eXplainable and eXtendable Deepfake Detectioncited by 35
  5. 2025FakeScope: Large Multimodal Expert Model for Transparent AI-Generated Image Forensicscited by 11
  6. 2024Measuring the Measurers: Quality Evaluation of Hallucination Benchmarks for Large Vision-Language Modelscited by 8
  7. 2025Semantic Visual Anomaly Detection and Reasoning in AI-Generated Imagescited by 8
  8. 2026Fake-HR1: Rethinking Reasoning of Vision Language Model for Synthetic Image Detectioncited by 8
  9. 2025Explainable Synthetic Image Detection through Diffusion Timestep Ensemblingcited by 7
  10. 2026EvoGuard: An Extensible Agentic RL-based Framework for Practical and Evolving AI-Generated Image Detectioncited by 6
  11. 2025PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detectioncited by 5
  12. 2025From Evidence to Verdict: An Agent-Based Forensic Framework for AI-Generated Image Detectioncited by 5
24 more
  1. 2025REVEAL: Reasoning-Enhanced Forensic Evidence Analysis for Explainable AI-Generated Image Detectioncited by 5
  2. 2025Locate-Then-Examine: Grounded Region Reasoning Improves Detection of AI-Generated Imagescited by 4
  3. 2026FakeVLM-R1: Internalizing Physical Laws via CoT for Synthetic Image Detectioncited by 4
  4. 2026Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detectioncited by 3
  5. 2026AgentFoX: LLM Agent-Guided Fusion with eXplainability for AI-Generated Image Detectioncited by 3
  6. 2024EditScout: Locating Forged Regions from Diffusion-based Edited Images with Multimodal LLMcited by 2
  7. 2025SAGA: Source Attribution of Generative AI Videoscited by 2
  8. 2026GRRE: Leveraging G-Channel Removed Reconstruction Error for Robust Detection of AI-Generated Imagescited by 1
  9. 2026Exploiting the Final Component of Generator Architectures for AI-Generated Image Detectioncited by 1
  10. 2026ArtifactLens: Hundreds of Labels Are Enough for Artifact Detection with VLMscited by 1
  11. 2026GenVideoLens: Where LVLMs Fall Short in AI-Generated Video Detection?cited by 1
  12. 2026XPlainVerse: A Million-Scale Benchmark for Explainable Deepfake Detectioncited by 1
  13. 2025Unmasking Synthetic Realities in Generative AI: A Comprehensive Review of Adversarially Robust Deepfake Detection Systemscited by 0
  14. 2025Don't Guess, Escalate: Towards Explainable Uncertainty-Calibrated AI Forensic Agentscited by 0
  15. 2026Generalizable Detection of AI Generated Images with Large Models and Fuzzy Decision Treecited by 0
  16. 2026Multi-Feature Fusion Approach for Generative AI Images Detectioncited by 0
  17. 2026Beyond Semantics: Uncovering the Physics of Fakes via Universal Physical Descriptors for Cross-Modal Synthetic Detectioncited by 0
  18. 2026SynMulti: Synthetic-to-Real Learning for Multimodal Video Understandingcited by 0
  19. 2026AI-Generated Images: What Humans and Machines See When They Look at the Same Imagecited by 0
  20. 2026Do CNNs Internally Represent Real and Fake Images Differently? A Hidden-Layer Analysiscited by 0
  21. 2026Frontier vision-language models have overtaken young adults at detecting AI-generated portraits -- but not their calibrationcited by 0
  22. 2026Mandela-Bench: Multimodal Models Remember Canonical Images Instead of Seeing Themcited by 0
  23. 2026From Sharp Eyes to Expert Mind: Internalizing Expert Knowledge in MLLMs for Tampered Text Detectioncited by 0
  24. 2026Evaluating the Evaluators: Diagnosing Large Multimodal Models for AI-Generated Image Assessmentcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanation
  2. 2025Seeing Before Reasoning: A Unified Framework for Generalizable and Explainable Fake Image Detection
  3. 2025Toward Generalizable Forgery Detection and Reasoning
  4. 2025Towards Explainable Fake Image Detection with Multi-Modal Large Language Models
  5. 2025ForenX: Towards Explainable AI-Generated Image Detection with Multimodal Large Language Models
  6. 2024FakeShield: Explainable Image Forgery Detection and Localization via Multi-modal Large Language Models
  7. 2026Fact or Fake? Assessing the Role of Deepfake Detectors in Multimodal Misinformation Detection
  8. 2025Explainable Detection of AI-Generated Images with Artifact Localization Using Faster-Than-Lies and Vision-Language Models for Edge Devices