DIALECT-COPA: Extending the Standard Translations of the COPA Causal Commonsense Reasoning Dataset to South Slavic Dialects

Nikola Ljubešić, Nada Galant, Sonja Benčina, Jaka Čibej, Stefan Milosavljević, Peter Rupnik, Taja Kuzman


Abstract
The paper presents new causal commonsense reasoning datasets for South Slavic dialects, based on the Choice of Plausible Alternatives (COPA) dataset. The dialectal datasets are built by translating by native dialect speakers from the English original and the corresponding standard translation. Three dialects are covered – the Cerkno dialect of Slovenian, the Chakavian dialect of Croatian and the Torlak dialect of Serbian. The datasets are the first resource for evaluation of large language models on South Slavic dialects, as well as among the first commonsense reasoning datasets on dialects overall. The paper describes specific challenges met during the translation process. A comparison of the dialectal datasets with their standard language counterparts shows a varying level of character-level, word-level and lexicon-level deviation of dialectal text from the standard datasets. The observed differences are well reproduced in initial zero-shot and 10-shot experiments, where the Slovenian Cerkno dialect and the Croatian Chakavian dialect show significantly lower results than the Torlak dialect. These results show also for the dialectal datasets to be significantly more challenging than the standard datasets. Finally, in-context learning on just 10 examples shows to improve the results dramatically, especially for the dialects with the lowest results.
Anthology ID:
2024.vardial-1.7
Volume:
Proceedings of the Eleventh Workshop on NLP for Similar Languages, Varieties, and Dialects (VarDial 2024)
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Yves Scherrer, Tommi Jauhiainen, Nikola Ljubešić, Marcos Zampieri, Preslav Nakov, Jörg Tiedemann
Venues:
VarDial | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
89–98
Language:
URL:
https://aclanthology.org/2024.vardial-1.7
DOI:
10.18653/v1/2024.vardial-1.7
Bibkey:
Cite (ACL):
Nikola Ljubešić, Nada Galant, Sonja Benčina, Jaka Čibej, Stefan Milosavljević, Peter Rupnik, and Taja Kuzman. 2024. DIALECT-COPA: Extending the Standard Translations of the COPA Causal Commonsense Reasoning Dataset to South Slavic Dialects. In Proceedings of the Eleventh Workshop on NLP for Similar Languages, Varieties, and Dialects (VarDial 2024), pages 89–98, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
DIALECT-COPA: Extending the Standard Translations of the COPA Causal Commonsense Reasoning Dataset to South Slavic Dialects (Ljubešić et al., VarDial-WS 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.vardial-1.7.pdf
Supplementary material:
 2024.vardial-1.7.SupplementaryMaterial.txt