(Apologies for cross-posting)

Dear colleagues,

We invite participants to a three-day winter school on large-scale NLP research, with special emphasis on pretraining data quality and multilingual evaluation of large language models (LLMs). The school will provide lectures and space for discussion by the following invited speakers:

- Jenia Jitsev (Jülich Supercomputing Centre) - Marianna Nezhurina (LAION) - Guilherme Penedo (Huggingface) - Gema Ramírez-Sánchez (Prompsit Language Engineering) - Anna Rogers (IT University of Copenhagen) - Ahmet Üstün (Cohere AI) - other international experts to be confirmed.

The winter school is organized as a collaboration between the Horizon Europe project High-Performance Language Technologies (HPLT, https://hplt-project.org/) and the Nordic Language Processing Laboratory (NLPL). The event will be held ‘in real life’ on February 3–5, 2025, in Norway. For additional information, please see:

http://wiki.nlpl.eu/Community/training

There is no participant fee for the winter school, and HPLT will provide free bus transfer between the Oslo airport and the conference hotel (about two hours north of Oslo, with skiing facilities just outside the door). Participants will need to cover their own travel to Oslo and accommodation at the hotel (NOK 3855 for two nights in a single room, including all meals and conference facilities).

We kindly invite expressions of interest in participation in the winter school. Please register through the on-line form linked up from the above overview page. We will process requests for participation on a first-come, first-served basis, with an eye toward regional balance. Participation will be confirmed in three batches, one on December 6, another one on December 13, and finally after the closing date for registration, which is Friday, December 20, 2024.

Welcome to Skeikampen in February 2025!

Andrey Kutuzov & Stephan Oepen (for the organizing team)