📢 PolEval 2025 Challenge Has Begun!
We are pleased to announce the launch of the 8th edition of the PolEval competition – an annual evaluation campaign for natural language processing (NLP) tools for Polish.
PolEval aims to improve the quality of existing solutions, test new algorithms, and promote linguistic engineering and AI research for the Polish language. Participants are invited to submit their NLP systems to compete in several shared tasks using standardized datasets and evaluation procedures. 🧩 PolEval 2025 Tasks
This year’s edition includes the following tasks:
-
ŚMIGIEL – Spotting Machine-Generated Text from Language Models for Polish
https://poleval.pl/tasks/task1 -
Gender-Inclusive LLMs for Polish
https://poleval.pl/tasks/task2 -
Polish Language Document Layout Detection
https://poleval.pl/tasks/task3 -
Polish Speech Emotion Recognition Challenge
https://poleval.pl/tasks/task4
📅 Important Dates
-
Submissions are now open! The deadline for submitting systems is November 10th, 2025.
🏆 Awards and Opportunities
Winners of individual tasks will receive awards, and all participants will be invited to:
-
Present their solutions during Data Science Summit 2025 https://main.dssconf.pl/, and
-
Publish their results in a dedicated PolEval 2025 proceedings volume.
🔗 More Information
-
Official website: http://poleval.pl/
-
Submission platform: https://poleval.amueval.pl/
We kindly invite all researchers, students, and enthusiasts of linguistic engineering to take part and spread the word about PolEval 2025 among potential participants. International participants are welcome, as all the information about the challenge is provided in English. With best regards, Łukasz Kobyliński, Maciej Ogrodniczuk, and the Task Organizers