%0 Conference Proceedings
%A List, Johann-Mattis
%A Vylomova, Ekaterina
%A Forkel, Robert
%A Hill, Nathan W.
%A Cotterell, Ryan D.
%+ Department of Linguistic and Cultural Evolution, Max Planck Institute for Evolutionary Anthropology, Max Planck Society
Department of Linguistic and Cultural Evolution, Max Planck Institute for Evolutionary Anthropology, Max Planck Society
Department of Linguistic and Cultural Evolution, Max Planck Institute for Evolutionary Anthropology, Max Planck Society
%T The SIGTYP 2022 shared task on the prediction of cognate reflexes : 
%G eng
%U https://hdl.handle.net/21.11116/0000-000B-F2B8-4
%R 10.18653/v1/2022.sigtyp-1.7
%D 2022
%B 4th Workshop on Computational Typology and
Multilingual NLP
%Z date of event: 2022-06-14 - 2022-06-14
%C Seattle, Washington
%X This study describes the structure and the results of the SIGTYP 2022 shared task on the prediction of cognate reflexes from multilingual wordlists. We asked participants to submit systems that would predict words in individual languages with the help of cognate words from related languages. Training and surprise data were based on standardized multilingual wordlists from several language families. Four<br>teams submitted a total of eight systems, in-<br>cluding both neural and non-neural systems,<br>as well as systems adjusted to the task and sys-<br>tems using more general settings. While all sys-<br>tems showed a rather promising performance,<br>reflecting the overwhelming regularity of sound<br>change, the best performance throughout was achieved by a system based on convolutional networks originally designed for image restoration.
%B SIGTYP 2022 : the 4th workshop on computational typology and
multilingual NLP : proceedings of the workshop
%E Vylomova, Ekaterina; Ponti, Edoardo; Cotterell, Ryan
%P 52 - 62
%I Association for Computational Linguistics (ACL)
%@ 978-1-955917-93-3