PolEval 2026 – Call for task proposals

Alina Wróblewska via Corpora <[email protected]>
Newsgroups gmane.science.linguistics.corpora
Message-ID <[email protected]>
Dear All,

At the turn of 2026 and 2027, the next edition of PolEval, a shared task competition for computational tools for processing Polish, will take place. The aim of the shared task is to improve the quality of existing solutions, test new algorithms and methods, and support the development of computational linguistics.

We are now opening the Call for task proposals for PolEval 2026. Until September 7, 2026, we invite submissions from individuals, teams, and organisations interested in preparing and coordinating shared tasks as part of the competition.

A task proposal should include:
task title – a short and unambiguous name;
task description – outlining its goal, scope, and main assumptions;
motivation and relevance – in particular, information on whether a similar task has already been organised as part of other competitions or benchmarks, e.g. for other languages; if the proposal concerns a new problem or a problem specific to Polish, please highlight its uniqueness and potential importance for the development of natural language processing;
task preparation status – in particular, information on the availability and status of training and test data, any licensing restrictions, and the expected date by which the complete dataset will be ready;
evaluation procedure – a description of how systems will be evaluated, including the proposed metrics, result submission procedure, etc.;
proposed timeline for running the task as part of PolEval 2026 – in particular, the expected dates for data release and completion of the evaluation.
Task organisers will be responsible for preparing the task description, data, and evaluation procedure; communicating with participants; coordinating the review process for submitted papers; and publishing a paper summarising the task in the PolEval proceedings in the ACL Anthology <https://aclanthology.org/venues/poleval/>.

Task proposals should be sent to: [email protected] <mailto:[email protected]>
PolEval 2026 schedule
Call for task proposals: August 11, 2026
Task submission deadline: September 7, 2026
Notification of task acceptance: September 14, 2026
First call for participation/papers & release of Train and Test A data: October 13, 2026
Second call for participation/papers: November 13, 2026
Release of Test B data: December 1, 2026
Announcement of final results (Test B): December 8, 2026
Paper submission deadline: December 15, 2026
Notification of paper acceptance: January 5, 2027
Camera-ready paper due: January 19, 2027
Presentation of the results and publication of the PolEval proceedings in the ACL Anthology: March 2027
The following tasks have been organised in previous editions of PolEval:

Part of Speech Tagging
Sentiment Analysis
Dependency Parsing
Named Entity Recognition
Language Models
Recognition and normalization of temporal expressions
Lemmatization of proper names and multi-word phrases
Entity linking
Machine translation
Automatic speech recognition
Automatic cyberbullying detection
Post-editing and rescoring of automatic speech recognition results
Morphosyntactic tagging of Middle, New and Modern Polish
Word sense disambiguation
Information extraction and entity typing from long documents with complex layouts
Punctuation restoration from read text
Evaluation of translation quality assessment metrics
Post-correction of OCR results
Question answering challenge
Punctuation prediction from conversational language
Abbreviation disambiguation
Passage retrieval
Reading Comprehension
Emotion and sentiment recognition
Polish Automatic Speech Recognition Challenge
Spotting Machine-Generated Text from Language Models for Polish (ŚMIGIEL)
Gender-inclusive LLMs for Polish
Polish Speech Emotion Recognition Challenge
Across these tasks, more than 140 systems have competed, developed by researchers from academia and by private companies. Organising a task is a great opportunity not only to test and compare existing solutions, but also to draw the attention of the research and engineering communities to important challenges in Polish language processing.

More information about the competition is available on:
the PolEval website <http://poleval.pl/>;
the PolEval Discord server <https://discord.gg/dpn94tUSyT>, used for ongoing discussions and information exchange;
PolEval on LinkedIn <https://www.linkedin.com/company/18152015>.
We would also greatly appreciate your help in spreading the word about PolEval 2026 among your colleagues, students, PhD students, and others interested in computational linguistics and NLP.

Thank you in advance for your support and involvement. We look forward to receiving interesting proposals for tasks for PolEval 2026!

Best regards,
Łukasz Kobyliński
Maciej Ogrodniczuk
Alina Wróblewska
Institute of Computer Science, Polish Academy of Sciences

_______________________________________________
Corpora mailing list -- [email protected]
https://list.elra.info/mailman3/postorius/lists/corpora.list.elra.info/
To unsubscribe send an email to [email protected]
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.