IberAuTexTification (test dataset)

Test datasets for the IberAuTexTification shared task at IberLEF 2024. This task aims to boost research on the detection and attribution of text generated automatically by large language models, in a multilingual (languages from the Iberian peninsula such as Spanish, English, Catalan, Gallego, Euskera, and Portuguese), multi-domain (news, reviews, emails, essays, dialogues, wikipedia, wikihow, tweets, emails, etc.), and multi-model (GPT, LLaMA, Mistral, Cohere, Anthropic, MPT, Falcon, etc.) setup.

This dataset includes only the test sets (without labels) for the two subtasks of the competition.

Once you request the data through Zenodo, a password to decompress the data will be sent to your email within 24 hours. Please, make sure you write your email address correctly, since we will send passwords, as well as future notifications to that address.