A Dynamic, Time-Aware Benchmark · ACL 2026

LiveFact

Can LLMs reason under the "fog of war," or just recite what they memorized?

LiveFact scores models against evidence sliced at three days before, on, and three days after each news event broke — using fresh events from Google News' World topic, refreshed every month so the test set is always unseen.

Running Months
Claims Evaluated
Models Evaluated

About LiveFact

A continuously updated benchmark that tests reasoning under real-time uncertainty — not memorized recall

Fog-of-War Testing

Every claim is evaluated at three time slices — using only the evidence that existed 3 days before, on, and 3 days after the event's headline date.

Real / Fake / Ambiguous

Claims aren't forced into a binary verdict. "Ambiguous" is a correct answer in its own right whenever the evidence genuinely can't support a verdict yet.

Dual-Mode Evaluation

Classification Mode checks factual accuracy against the final, time-invariant truth. Inference Mode checks whether a model knows when it doesn't know yet.

Monthly, World-Only Refresh

LiveFact draws exclusively from Google News' World topic and refreshes every month with unseen events, keeping every release a genuine zero-shot test.

T − 3 days Little confirming evidence yet — correct answer is usually "Ambiguous"
T · headline day Evidence starts to confirm Real or Fake
T + 3 days Evidence has settled — the verdict is clear

Leaderboard

Current rankings of LLMs on the LiveFact benchmark

Every claim is sourced from Google News' World section, refreshed monthly.

November 2025
Rank Model Organization Overall Avg. Before Event (CLS) Before Event (INF) During Event (CLS) During Event (INF) After Event (CLS) After Event (INF)

Methodology

How each monthly LiveFact release is built

01

Event Retrieval

We scrape trending stories from Google News' World topic daily at 00:00 GMT, then deduplicate and filter down to a core set of distinct, high-impact events.

02

Time-Sliced Evidence

For every event we build three evidence sets anchored to its headline date — T−3, T, and T+3 — simulating the real-world "fog of war" of incomplete information.

03

Claim & Context Generation

An LLM drafts neutral background context plus a balanced set of Real, Fake, and Ambiguous claims for each event, grounded in the retrieved evidence.

04

Human-in-the-Loop Verification

Every claim and label goes through independent expert review rounds against the source evidence before it's accepted into the release.

05

Dual-Mode Scoring

Models are scored in Classification Mode (fixed ground truth) and Inference Mode (evidence-adjusted ground truth) at each of the three time slices.

Get Involved

Join our community and contribute to advancing fake news detection research

Submit Your Model

Have a model you'd like to see on the leaderboard? Submit it for evaluation.

Submit Model

Access Dataset

Download the LiveFact dataset and evaluation scripts for your research on Hugging Face.

Download Dataset

Contribute

Join the LiveFact maintenance team and help run the ongoing monthly benchmark cycle.

Join the Team

Cite This Work

If LiveFact is useful in your research, please cite our ACL 2026 paper.

@inproceedings{xu2026livefact,
    title = "{L}ive{F}act: A Dynamic, Time-Aware Benchmark for {LLM}-Driven Fake News Detection",
    author = "Xu, Cheng  and  Jin, Changhong  and  Niu, Yingjie  and  Yan, Nan  and  Mei, Yuke  and  Guan, Shuhao  and  Chen, Liming  and  Kechadi, Tahar",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.acl-long.546/",
    doi = "10.18653/v1/2026.acl-long.546",
    pages = "11881--11910"
}