🏆 Best Paper Award at WaC-13!

The TurkuNLP team had a successful week at EMNLP 2026, held in Budapest on 24–29 October. At the conference’s 13th Web-as-Corpus (WaC-13), Erik Henriksson and colleagues received the Best Paper Award for the paper:

“Segment-Level Register Analysis of the Multilingual Web” by Erik Henriksson, Alireza Razzaghi, Tomasz Galica, Jiaqi Feng Guo, Anastasia Zhi-Ling Dong, Urho Kunnas, Erofili Psaltaki, Tuomas Lundberg, Antti Kanner and Veronika Laippala.

The team also presented several other papers at WaC-13 and EMNLP, showcasing research on multilingual web data, register analysis, and document quality:

–  “HPLT3-PerRef: A Refined Large-Scale Persian Web Corpus via Supervised Document Quality Filtering” by Alireza Razzaghi, Tomasz Galica, Erik Henriksson and Veronika Laippala

– “Is Educational Content Enough? The Linguistic and Thematic Effects of Educational Filtering” by Urho Kunnas, Veronika Laippala, Sampo Pyysalo and Tomasz Galica

–  “Towards More Reliable Web Metadata: Do Register Classifiers Learn Register or Genre Features?” by Antti Kanner, Tuomas Lundberg, Erik Henriksson and Veronika Laippala

– “Task-Agnostic Web Document Annotation with LLM-Generated Descriptors” by Otto Tarkka, Erik Henriksson, Jenna Kanerva and Filip Ginter

– “Register-based Data Selection for Multilingual LLMs” by Tomasz Galica, Amanda Myntti, Erik Henriksson, Fedor Vitiugin, Sampo Pyysalo and Veronika Laippala.