CheckThat! Lab at CLEF 2025

Home

Editions

Tasks

Contents

Task 2: Claims Extraction & Normalization

Definition

With an upsurge in social media, users have been exposed to many misleading claims. However, the ubiquitous noise in these posts makes it difficult for manual fact-checkers to quickly identify specific and precise claims that need verification. Given a noisy, unstructured social media post, the task is to simplify it into a concise form. Therefore, we aim to bridge this gap by decomposing social media posts into simpler, more comprehensible forms, which are referred to as normalized claims.

‘‘Given a noisy, unstructured social media post, the task is to simplify it into a concise form.’’

The task is a generation task offered in 20 languages: English, Arabic, Bengali, Czech, German, Greek, French, Hindi, Korean, Marathi, Indonesian, Dutch, Punjabi, Polish, Portuguese, Romanian, Spanish, Tamil, Telugu, Thai.

Datasets

Training and test datasets will be released on GitLab.

Evaluation

We will employ METEOR score for final evaluation.

Submission

TBA

Leaderboard

TBA

Organizers

  • Preslav Nakov, MBZUAI, UAE
  • Tanmoy Chakraborty, Indian Institute of Technology Delhi, India
  • Megha Sundriyal, Indraprastha Institute of Information Technology Delhi, India

Contact

For queries, please join the Slack channel

Alternatively, please send an email to: clef-factcheck@googlegroups.com