Proceedings of Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2023)
These proceedings include the 23 papers presented at the 10th Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial), co-located with the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Both EACL and VarDial were held in Dubrovnik, Croatia, in a hybrid format, allowing participants to attend on-site or to participate virtually. This edition marks VarDial’s ten-year anniversary. We are pleased to see that the workshop continues to serve the community as the main venue for researchers interested in the computational processing of diatopic language variation. The papers accepted this year address a wide range of topics, such as corpus building, part-of-speech tagging, and machine translation. This volume once again showcases the great linguistic diversity that VarDial embodies, including work on dialects and varieties of many different languages, such as Arabic, Cantonese, Croatian, Finnish, German, Irish, Italian, Mandarin, Occitan, Serbian, and Spanish. The VarDial evaluation campaign continues to be an essential part of the workshop. In VarDial 2023, three shared tasks were organized: Slot and intent detection for low-resource language varieties (SID4LR), Discriminating Between Similar Languages – True Labels (DSL-TL), and Discriminating Between Similar Languages – Speech (DSL-S). All three tasks were organized for the first time this year. This volume includes the system description papers prepared by the participating teams, as well as a report written by the task organizers summarizing the results and the findings of the evaluation campaign. Finally, we would like to take this opportunity to thank all the shared task organizers and the participants for their hard work. We further thank the VarDial program committee members for being an important part of the workshop’s success over these ten years.