On Evaluating Multilingual Compositional Generalization with Translated Datasets
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
On Evaluating Multilingual Compositional Generalization with Translated Datasets. / Wang, Zi; Hershcovich, Daniel.
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics (ACL), 2023. p. 1669-1687.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - On Evaluating Multilingual Compositional Generalization with Translated Datasets
AU - Wang, Zi
AU - Hershcovich, Daniel
N1 - Publisher Copyright: © 2023 Association for Computational Linguistics.
PY - 2023
Y1 - 2023
N2 - Compositional generalization allows efficient learning and human-like inductive biases. Since most research investigating compositional generalization in NLP is done on English, important questions remain underexplored. Do the necessary compositional generalization abilities differ across languages? Can models compositionally generalize cross-lingually? As a first step to answering these questions, recent work used neural machine translation to translate datasets for evaluating compositional generalization in semantic parsing. However, we show that this entails critical semantic distortion. To address this limitation, we craft a faithful rule-based translation of the MCWQ dataset (Cui et al., 2022) from English to Chinese and Japanese. Even with the resulting robust benchmark, which we call MCWQ-R, we show that the distribution of compositions still suffers due to linguistic divergences, and that multilingual models still struggle with cross-lingual compositional generalization. Our dataset and methodology will be useful resources for the study of cross-lingual compositional generalization in other tasks.
AB - Compositional generalization allows efficient learning and human-like inductive biases. Since most research investigating compositional generalization in NLP is done on English, important questions remain underexplored. Do the necessary compositional generalization abilities differ across languages? Can models compositionally generalize cross-lingually? As a first step to answering these questions, recent work used neural machine translation to translate datasets for evaluating compositional generalization in semantic parsing. However, we show that this entails critical semantic distortion. To address this limitation, we craft a faithful rule-based translation of the MCWQ dataset (Cui et al., 2022) from English to Chinese and Japanese. Even with the resulting robust benchmark, which we call MCWQ-R, we show that the distribution of compositions still suffers due to linguistic divergences, and that multilingual models still struggle with cross-lingual compositional generalization. Our dataset and methodology will be useful resources for the study of cross-lingual compositional generalization in other tasks.
UR - http://www.scopus.com/inward/record.url?scp=85174398526&partnerID=8YFLogxK
U2 - 10.18653/v1/2023.acl-long.93
DO - 10.18653/v1/2023.acl-long.93
M3 - Article in proceedings
AN - SCOPUS:85174398526
SP - 1669
EP - 1687
BT - Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
PB - Association for Computational Linguistics (ACL)
T2 - 61st Annual Meeting of the Association for Computational Linguistics, ACL 2023
Y2 - 9 July 2023 through 14 July 2023
ER -
ID: 372526809