TrustworthyComp/ClaimReview2025Q4

ClaimReview 2025Q4

Links

• 📑 ArXiv: https://arxiv.org/abs/2607.23514
• 💻 Code: https://github.com/TrustworthyComp/Rethink-MAFC-Eval
• 🤗 HF Daily Paper: https://huggingface.co/papers/2607.23514

ClaimReview 2025Q4 is a curated benchmark of recent fact-checking claims and verdicts collected from IFCN-aligned fact-checking organizations.
This release contains 901 English claim-verification instances published between 2025-10-01 and 2025-12-31, and is designed for evaluating automated fact-checking systems under a dynamic, time-sensitive setting.

NEWS

• 2026-07-26 📄 – Preprint released on arXiv.
• 2026-07-23 🏆 – Accepted by MM-2026; see you in Brazil! 🇧🇷
• 2026-07-15 🎉 – Source code is publicly released on GitHub; contributions and feedback are welcome!

Dataset

The dataset file is:

• claim_review_2025q4.json

Each instance follows the structure below:

{
  "claim": "The claim text itself.",
  "label": "One of: SUPPORTED, REFUTED, NEI.",
  "review_url": "URL of the source fact-check article.",
  "fact_checker": {
    "name": "Fact-checking organization name.",
    "country": "Country of the fact-checking organization.",
    "language": "Language of the fact-check article.",
    "website": "Official website of the organization.",
    "ifcn_url": "IFCN profile URL of the organization.",
    "avatar": "IFCN-hosted avatar image URL.",
    "domain": "Primary web domain of the organization."
  },
  "date_published": "Publication date of the fact-check article (YYYY-MM-DD)."
}

Dataset Statistics

• Total instances: 901
• Label distribution:
  • REFUTED: 606
  • NEI: 169
  • SUPPORTED: 126
• Fact-checking organizations: 14
• Countries covered: 7
• Date range: 2025-10-01 to 2025-12-31

Label Definitions

• SUPPORTED: The claim is supported by the arguments and evidence presented.
• REFUTED: The claim is contradicted by the arguments and evidence presented.
• NEI: The presented evidence is not enough to support or refute the claim. It applies when the evidence either explicitly indicates that relevant evidence cannot be found or leaves certain aspects of the claim neither supported nor refuted.

Intended Uses

This dataset is used in our paper for sourcing up-to-date fact-checks:
Novel Claim or Déjà Vu? Rethinking "Contamination-Free" Dynamic Evaluation for Multimodal Automated Fact-Checking.

Data Quality Notes

• All records include all top-level fields shown in the schema above.
• The dataset aggregates claims from multiple professional fact-checking outlets.
• URLs and publisher metadata are preserved to support traceability and evidence lookup.

Quick Start

import json

with open("claim_review_2025q4.json", "r", encoding="utf-8") as f:
    data = json.load(f)

print(f"Loaded {len(data)} instances")
print(data[0].keys())

Citation

If you use this dataset in your work, please cite:

@inproceedings{rethink_mafc_eval_2026,
  title={Novel Claim or Déjà Vu? Rethinking "Contamination-Free" Dynamic Evaluation for Multimodal Automated Fact-Checking},
  author={He, Haorui and Chen, Xinwen and Wen, Dacheng and Cheng, Reynold and Lau, Francis C. M. and Li, Yupeng},
  booktitle={Proc.~of MM},
  year={2026}
}

License

This dataset card is released under CC BY-NC 4.0.
Please make sure your downstream use also complies with the terms and with the usage policies of the original fact-check sources.