DEFAME: Dynamic Evidence-based FAct-checking with Multimodal Experts
Not on the map. Why?
Abstract (arXiv)
The proliferation of disinformation demands reliable and scalable fact-checking solutions. We present Dynamic Evidence-based FAct-checking with Multimodal Experts (DEFAME), a modular, zero-shot MLLM pipeline for open-domain, text-image claim verification. DEFAME operates in a six-stage process, dynamically selecting the tools and search depth to extract and evaluate textual and visual evidence. Unlike prior approaches that are text-only, lack explainability, or rely solely on parametric knowledge, DEFAME performs end-to-end verification, accounting for images in claims and evidence while generating structured, multimodal reports. Evaluation on the popular benchmarks VERITE, AVerITeC, and MOCHEG shows that DEFAME surpasses all previous methods, establishing itself as the new state-of-the-art fact-checking system for uni- and multimodal fact-checking. Moreover, we introduce a new multimodal benchmark, ClaimReview2024+, featuring claims after the knowledge cutoff of GPT-4o, avoiding data leakage. Here, DEFAME drastically outperforms the GPT-4o baselines, showing temporal generalizability and the potential for real-time fact-checking.
Key facts. Cited by 43 papers in the map (data through 14 Oct 2026). Builds on 64 of its 92 references that are in the map. First posted on arXiv 13 Dec 2024.
Builds on
64 of 92 references.
References · 64
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
- 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
- 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,375
- 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,071
- 2020On Faithfulness and Factuality in Abstractive Summarizationcited by 900
- 2023Bias and Fairness in Large Language Models: A Surveycited by 639
- 2023ViperGPT: Visual Inference via Python Execution for Reasoningcited by 603
- 2022Socratic Models: Composing Zero-Shot Multimodal Reasoning with Languagecited by 513
- 2017"Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detectioncited by 363
- 2023GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localizationcited by 187
- 2023Can LLM-Generated Misinformation Be Detected?cited by 177
- 2023Fact-Checking Complex Claims with Program-Guided Reasoningcited by 146
38 more
- 2023AVeriTeC: A Dataset for Real-world Claim Verification with Evidence from the Webcited by 144
- 2020Generating Fact Checking Explanationscited by 132
- 2022TruFor: Leveraging all-round clues for trustworthy image forgery detection and localizationcited by 130
- 2021NewsCLIPpings: Automatic Generation of Out-of-Context Multimodal Mediacited by 107
- 2022End-to-End Multimodal Fact-Checking and Explanation Generation: A Challenging Dataset and Modelscited by 107
- 2023Detecting and Grounding Multi-Modal Media Manipulationcited by 99
- 2023Task Contamination: Language Models May Not Be Few-Shot Anymorecited by 99
- 2023Towards LLM-based Fact Verification on News Claims with a Hierarchical Step-by-Step Prompting Methodcited by 97
- 2023Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkerscited by 87
- 2024SNIFFER: Multimodal Large Language Model for Explainable Out-of-Context Misinformation Detectioncited by 74
- 2021Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resourcescited by 71
- 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensicscited by 69
- 2022MuMiN: A Large-Scale Multilingual Multimodal Fact-Checked Misinformation Social Network Datasetcited by 61
- 2024MMFakeBench: A Mixed-Source Multimodal Misinformation Detection Benchmark for LVLMscited by 61
- 2023Multimodal Automated Fact-Checking: A Surveycited by 53
- 2019Fact-Checking Meets Fauxtography: Verifying Claims About Imagescited by 52
- 2018On the Origins of Memes by Means of Fringe Web Communitiescited by 47
- 2023Benchmarks for Automated Commonsense Reasoning: A Surveycited by 40
- 2023VERITE: A Robust Benchmark for Multimodal Misinformation Detection Accounting for Unimodal Biascited by 39
- 2021AmbiFC: Fact-Checking Ambiguous Claims with Evidencecited by 38
- 2023FactLLaMA: Optimizing Instruction-Following Language Models with External Knowledge for Automated Fact-Checkingcited by 37
- 2023Reading and Reasoning over Chart Images for Evidence-based Automated Fact-Checkingcited by 35
- 2023Charting New Territories: Exploring the Geographic and Geospatial Capabilities of Multimodal LLMscited by 34
- 2020Multimodal Analytics for Real-world News using Measures of Cross-modal Entity Consistencycited by 33
- 2018Deep Multimodal Image-Repurposing Detectioncited by 31
- 2023Self-Checker: Plug-and-Play Modules for Fact-Checking with Large Language Modelscited by 31
- 2024The Automated Verification of Textual Claims (AVeriTeC) Shared Taskcited by 29
- 2017Multimedia Semantic Integrity Assessment Using Joint Embedding Of Images And Textcited by 28
- 2024OpenFactCheck: A Unified Framework for Factuality Evaluation of LLMscited by 28
- 2024RAGAR, Your Falsehood Radar: RAG-Augmented Reasoning for Political Fact-Checking using Multimodal Large Language Modelscited by 27
- 2024AMMeBa: A Large-Scale Survey and Dataset of Media-Based Misinformation In-The-Wildcited by 27
- 2023ChartCheck: Explainable Fact-Checking over Real-World Chart Imagescited by 26
- 2024Claim Verification in the Age of Large Language Models: A Surveycited by 24
- 2024Multimodal Misinformation Detection using Large Vision-Language Modelscited by 21
- 2024Generative Large Language Models in Automated Fact-Checking: A Surveycited by 19
- 2023RED-DOT: Multimodal Fact-checking via Relevant Evidence Detectioncited by 18
- 2024Re-Search for The Truth: Multi-round Retrieval-augmented Large Language Models are Strong Fake News Detectorscited by 16
- 2024Improving Faithfulness of Large Language Models in Summarization via Sliding Generation and Self-Consistencycited by 14
50 most cited of 64 All 64 (JSON)
Built on by
Citing papers per year: 2025: 19; 2026: 24.
43 papers.
Sort:
By area:
- 2025AVerImaTeC: A Dataset for Automatic Verification of Image-Text Claims with Evidence from the Webcited by 18
- 2025COVE: COntext and VEracity prediction for out-of-context imagescited by 14
- 2025Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoningcited by 14
- 2025M4FC: a Multimodal, Multilingual, Multicultural, Multitask Real-World Fact-Checking Datasetcited by 7
- 2025Debating Truth: Debate-driven Claim Verification with Multiple Large Language Model Agentscited by 6
- 2025Drifting Away from Truth: GenAI-Driven News Diversity Challenges LVLM-Based Misinformation Detectioncited by 6
- 2026VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checkingcited by 6
- 2025Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Modelscited by 5
- 2025Adversarial Attacks Against Automated Fact-Checking: A Surveycited by 5
- 2025T^2Agent A Tool-augmented Multimodal Misinformation Detection Agent with Monte Carlo Tree Searchcited by 4
- 2025Multimedia Verification Through Multi-Agent Deep Research Multimodal Large Language Modelscited by 4
- 2025Fact2Fiction: Targeted Poisoning Attack to Agentic Fact-checking Systemcited by 4
31 more
- 2025VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrievalcited by 3
- 2025Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformationcited by 3
- 2025PASS-FC: Progressive and Adaptive Search Scheme for Fact Checking of Comprehensive Claimscited by 2
- 2026DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systemscited by 2
- 2026RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wildcited by 2
- 2025Beyond Retrieval: Improving Evidence Quality for LLM-based Multimodal Fact-Checkingcited by 1
- 2025MultiCheck: Strengthening Web Trust with Unified Multimodal Fact Verificationcited by 1
- 2026KG-ViP: Bridging Knowledge Grounding and Visual Perception in Multi-modal LLMs for Visual Question Answeringcited by 1
- 2026CaseFacts: A Benchmark for Legal Fact-Checking and Precedent Retrievalcited by 1
- 2026Fact or Fake? Assessing the Role of Deepfake Detectors in Multimodal Misinformation Detectioncited by 1
- 2026Multimodal Claim Extraction for Fact-Checkingcited by 1
- 2026When Misinformation Speaks and Converses: Rethinking Fact-Checking in Audio Platformscited by 1
- 2025ClaimCheck: Real-Time Fact-Checking with Small Language Modelscited by 0
- 2025Towards Real-Time Fake News Detection under Evidence Scarcitycited by 0
- 2025RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wildcited by 0
- 2026Multilingual, Multimodal Pipeline for Creating Authentic and Structured Fact-Checked Claim Datasetcited by 0
- 2026Frame-Guided Synthetic Claim Generation for Automatic Fact-Checking Using High-Volume Tabular Datacited by 0
- 2026Ostrakon-VL: Towards Domain-Expert MLLM for Food-Service and Retail Storescited by 0
- 2026Temporal Leakage in Search-Engine Date-Filtered Web Retrieval: A Retrospective Forecasting Case Studycited by 0
- 2026D-SECURE: Dual-Source Evidence Combination for Unified Reasoning in Misinformation Detectioncited by 0
- 2026Is a Picture Worth a Thousand Words? Adaptive Multimodal Fact-Checking with Visual Evidence Necessitycited by 0
- 2026Taxonomy of Risks on Automated Fact-Checking Systems Considering its Propagationcited by 0
- 2026Evidence Triangulation for Multimodal Fact-Checking in the Wildcited by 0
- 2026VETO: Towards Protecting Images From Frontier AI Editingcited by 0
- 2026How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluationcited by 0
- 2026ElementCheck: Complexity-Aware Long-Form Text Factuality Evaluation via Sentence Elementscited by 0
- 2026Evaluating and Improving Evidence-Grounded Fact-Checking in LLMs via Multi-Round Evidence Ablationcited by 0
- 2026Self-Evolving Multimedia Verification through Memory Consolidation of Contestation Experiencescited by 0
- 2026MM-VeriAgent: Learning to Use Extensive Tools to Verify Multimodal Misinformation with Reinforcement Learningcited by 0
- 2026OpenFC: Learning Verification Policies towards Open-Search Fact Checkingcited by 0
- 2026EviScene: Beyond "Is It AI?" for Agentic Image Factuality Assessmentcited by 0
- 2026EviScene: Beyond "Is It AI?" for Agentic Image Factuality Assessment10 Oct 2026
- 2026OpenFC: Learning Verification Policies towards Open-Search Fact Checking27 Sep 2026
- 2026MM-VeriAgent: Learning to Use Extensive Tools to Verify Multimodal Misinformation with Reinforcement Learning25 Sep 2026
- 2026Self-Evolving Multimedia Verification through Memory Consolidation of Contestation Experiences23 Sep 2026
- 2026Evaluating and Improving Evidence-Grounded Fact-Checking in LLMs via Multi-Round Evidence Ablation8 Sep 2026
- 2026How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation26 Aug 2026
- 2026VETO: Towards Protecting Images From Frontier AI Editing29 Jul 2026
- 2026Evidence Triangulation for Multimodal Fact-Checking in the Wild30 Jun 2026
- 2026Taxonomy of Risks on Automated Fact-Checking Systems Considering its Propagation24 Jun 2026
- 2026ElementCheck: Complexity-Aware Long-Form Text Factuality Evaluation via Sentence Elements17 Jun 2026
- 2026RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild11 May 2026
- 2026When Misinformation Speaks and Converses: Rethinking Fact-Checking in Audio Platforms18 Apr 2026
31 more
- 2026Is a Picture Worth a Thousand Words? Adaptive Multimodal Fact-Checking with Visual Evidence Necessity6 Apr 2026
- 2026D-SECURE: Dual-Source Evidence Combination for Unified Reasoning in Misinformation Detection16 Feb 2026
- 2026Fact or Fake? Assessing the Role of Deepfake Detectors in Multimodal Misinformation Detection2 Feb 2026
- 2026Multimodal Claim Extraction for Fact-Checking1 Feb 2026
- 2026DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systems31 Jan 2026
- 2026Temporal Leakage in Search-Engine Date-Filtered Web Retrieval: A Retrospective Forecasting Case Study31 Jan 2026
- 2026Ostrakon-VL: Towards Domain-Expert MLLM for Food-Service and Retail Stores29 Jan 2026
- 2026Frame-Guided Synthetic Claim Generation for Automatic Fact-Checking Using High-Volume Tabular Data23 Jan 2026
- 2026CaseFacts: A Benchmark for Legal Fact-Checking and Precedent Retrieval23 Jan 2026
- 2026KG-ViP: Bridging Knowledge Grounding and Visual Perception in Multi-modal LLMs for Visual Question Answering14 Jan 2026
- 2026VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checking13 Jan 2026
- 2026Multilingual, Multimodal Pipeline for Creating Authentic and Structured Fact-Checked Claim Dataset12 Jan 2026
- 2025RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild28 Dec 2025
- 2025M4FC: a Multimodal, Multilingual, Multicultural, Multitask Real-World Fact-Checking Dataset27 Oct 2025
- 2025Towards Real-Time Fake News Detection under Evidence Scarcity13 Oct 2025
- 2025ClaimCheck: Real-Time Fact-Checking with Small Language Models22 Sep 2025
- 2025Adversarial Attacks Against Automated Fact-Checking: A Survey10 Sep 2025
- 2025Drifting Away from Truth: GenAI-Driven News Diversity Challenges LVLM-Based Misinformation Detection18 Aug 2025
- 2025Fact2Fiction: Targeted Poisoning Attack to Agentic Fact-checking System8 Aug 2025
- 2025MultiCheck: Strengthening Web Trust with Unified Multimodal Fact Verification7 Aug 2025
- 2025Debating Truth: Debate-driven Claim Verification with Multiple Large Language Model Agents25 Jul 2025
- 2025Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformation9 Jul 2025
- 2025Multimedia Verification Through Multi-Agent Deep Research Multimodal Large Language Models6 Jul 2025
- 2025VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrieval26 May 2025
- 2025T^2Agent A Tool-augmented Multimodal Misinformation Detection Agent with Monte Carlo Tree Search26 May 2025
- 2025AVerImaTeC: A Dataset for Automatic Verification of Image-Text Claims with Evidence from the Web23 May 2025
- 2025Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoning22 May 2025
- 2025Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Models21 May 2025
- 2025Beyond Retrieval: Improving Evidence Quality for LLM-based Multimodal Fact-Checking6 May 2025
- 2025PASS-FC: Progressive and Adaptive Search Scheme for Fact Checking of Comprehensive Claims14 Apr 2025
- 2025COVE: COntext and VEracity prediction for out-of-context images3 Feb 2025
- 2025Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Modelscited by 5
- 2025Multimedia Verification Through Multi-Agent Deep Research Multimodal Large Language Modelscited by 4
- 2026KG-ViP: Bridging Knowledge Grounding and Visual Perception in Multi-modal LLMs for Visual Question Answeringcited by 1
- 2026Ostrakon-VL: Towards Domain-Expert MLLM for Food-Service and Retail Storescited by 0
- 2026Ostrakon-VL: Towards Domain-Expert MLLM for Food-Service and Retail Stores29 Jan 2026
- 2026KG-ViP: Bridging Knowledge Grounding and Visual Perception in Multi-modal LLMs for Visual Question Answering14 Jan 2026
- 2025Multimedia Verification Through Multi-Agent Deep Research Multimodal Large Language Models6 Jul 2025
- 2025Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Models21 May 2025
- 2025AVerImaTeC: A Dataset for Automatic Verification of Image-Text Claims with Evidence from the Webcited by 18
- 2025COVE: COntext and VEracity prediction for out-of-context imagescited by 14
- 2025Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoningcited by 14
- 2025M4FC: a Multimodal, Multilingual, Multicultural, Multitask Real-World Fact-Checking Datasetcited by 7
- 2025Drifting Away from Truth: GenAI-Driven News Diversity Challenges LVLM-Based Misinformation Detectioncited by 6
- 2026VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checkingcited by 6
- 2025Adversarial Attacks Against Automated Fact-Checking: A Surveycited by 5
- 2025T^2Agent A Tool-augmented Multimodal Misinformation Detection Agent with Monte Carlo Tree Searchcited by 4
- 2025VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrievalcited by 3
- 2025PASS-FC: Progressive and Adaptive Search Scheme for Fact Checking of Comprehensive Claimscited by 2
- 2026DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systemscited by 2
- 2026RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wildcited by 2
12 most cited of 34 All 34 (JSON)
- 2026OpenFC: Learning Verification Policies towards Open-Search Fact Checking27 Sep 2026
- 2026Self-Evolving Multimedia Verification through Memory Consolidation of Contestation Experiences23 Sep 2026
- 2026Evaluating and Improving Evidence-Grounded Fact-Checking in LLMs via Multi-Round Evidence Ablation8 Sep 2026
- 2026How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation26 Aug 2026
- 2026VETO: Towards Protecting Images From Frontier AI Editing29 Jul 2026
- 2026Evidence Triangulation for Multimodal Fact-Checking in the Wild30 Jun 2026
- 2026Taxonomy of Risks on Automated Fact-Checking Systems Considering its Propagation24 Jun 2026
- 2026ElementCheck: Complexity-Aware Long-Form Text Factuality Evaluation via Sentence Elements17 Jun 2026
- 2026RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild11 May 2026
- 2026When Misinformation Speaks and Converses: Rethinking Fact-Checking in Audio Platforms18 Apr 2026
- 2026Is a Picture Worth a Thousand Words? Adaptive Multimodal Fact-Checking with Visual Evidence Necessity6 Apr 2026
- 2026D-SECURE: Dual-Source Evidence Combination for Unified Reasoning in Misinformation Detection16 Feb 2026
12 newest of 34 All 34 (JSON)
Similar, not cited
Close in wording, no citation either way
- 2025HintsOfTruth: A Multimodal Checkworthiness Detection Dataset with Real and Synthetic Claims
- 2025MuSciClaims: Multimodal Scientific Claim Verification
- 2024Robust Claim Verification Through Fact Detection
- 2025Fact in Fragments: Deconstructing Complex Claims via LLM-based Atomic Fact Extraction and Verification
- 2026Checking Fact with Better Retrieval: Dynamic Contrastive Learning for Evidence Retrieval
- 2026Novel Claim or D'ej`a Vu? Rethinking "Contamination-Free'' Dynamic Evaluation for Multimodal Automated Fact-Checking
- 2024Piecing It All Together: Verifying Multi-Hop Multimodal Claims
- 2026MIC: Explaining Image-Claim Inconsistencies in AI-Generated Multimodal Misinformation