Multilingual Fact-Checking at Scale: Fine-Tuned Compact Models vs LLMs
Not on the map. Why?
Abstract (arXiv)
We present a multilingual fact-checking system deployed at Factiverse, designed for high-throughput and low-latency operation across diverse languages. The system follows a modular pipeline with three stages: claim detection, evidence retrieval and re-ranking, and veracity prediction. We fine-tune XLM-RoBERTa-Large for claim detection, mmBERT-base for three-label stance classification (Supports/Refutes/Mixed), and a SetFit-based multilingual re-ranker for claim--evidence matching. We compare these components against strong LLM baselines, including GPT-5.2, Claude Opus~4.6, and Qwen3-8b. Experiments on production data spanning 114 languages for claim detection and 28 languages for veracity prediction show that task-specific fine-tuning provides strong and stable multilingual performance, while the fine-tuned retrieval model remains competitive with modern proprietary embeddings. Same-hardware latency measurements further show large efficiency gains for encoder-based components, supporting their use in production deployments with tight cost and privacy constraints. Overall, compact fine-tuned, self-hosted models remain a practical and effective foundation for multilingual fact-checking at scale. Code and data used for this study are available at https://github.com/factiverse/factcheck-editor.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 24 of its 28 references that are in the map. First posted on arXiv 7 Jun 2026.
Builds on
24 of 28 references.
References · 24
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
- 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
- 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
- 2021A Survey on Automated Fact-Checkingcited by 378
- 2022RARR: Researching and Revising What Language Models Say, Using Language Modelscited by 287
- 2023FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarioscited by 251
- 2024Fine-grained Hallucination Detection and Editing for Language Modelscited by 149
- 2023AVeriTeC: A Dataset for Real-world Claim Verification with Evidence from the Webcited by 144
- 2022Efficient Few-Shot Learning Without Promptscited by 137
- 2018DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learningcited by 100
- 2023Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkerscited by 87
- 2020Fighting the COVID-19 Infodemic: Modeling the Perspective of Journalists, Fact-Checkers, Social Media Platforms, Policy Makers, and the Societycited by 72
12 more
- 2018A Survey on Natural Language Processing for Fake News Detectioncited by 67
- 2024Claim Detection for Automated Fact-checking: A Survey on Monolingual, Multilingual and Cross-Lingual Researchcited by 35
- 2019Jointly embedding the local and global relations of heterogeneous graph for rumor detectioncited by 31
- 2023The Perils & Promises of Fact-checking with Large Language Modelscited by 31
- 2020BRENDA: Browser Extension for Fake News Detectioncited by 20
- 2024Generative Large Language Models in Automated Fact-Checking: A Surveycited by 19
- 2017Finding Streams in Knowledge Graphs to Support Fact Checkingcited by 17
- 2024Surprising Efficacy of Fine-Tuned Transformers for Fact-Checking over Larger Language Modelscited by 14
- 2024FactCheck Editor: Multilingual Text Editor with End-to-End fact-checkingcited by 12
- 2024LiveFC: A System for Live Fact-Checking of Audio Streamscited by 5
- 2025FactIR: A Real-World Zero-shot Open-Domain Retrieval Benchmark for Fact-Checkingcited by 3
- 2025ShortCheck: Checkworthiness Detection of Multilingual Short-Form Videoscited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.