Multi-lingual and Multi-cultural Figurative Language Understanding
Document Type
Conference Proceeding
Publication Title
Proceedings of the Annual Meeting of the Association for Computational Linguistics
Abstract
Figurative language permeates human communication, but at the same time is relatively understudied in NLP. Datasets have been created in English to accelerate progress towards measuring and improving figurative language processing in language models (LMs). However, the use of figurative language is an expression of our cultural and societal experiences, making it difficult for these phrases to be universally applicable. In this work, we create a figurative language inference dataset, MABL, for seven diverse languages associated with a variety of cultures: Hindi, Indonesian, Javanese, Kannada, Sundanese, Swahili and Yoruba. Our dataset reveals that each language relies on cultural and regional concepts for figurative expressions, with the highest overlap between languages originating from the same region. We assess multilingual LMs' abilities to interpret figurative language in zero-shot and few-shot settings. All languages exhibit a significant deficiency compared to English, with variations in performance reflecting the availability of pre-training and fine-tuning data, emphasizing the need for LMs to be exposed to a broader range of linguistic and cultural variation during training.
First Page
8269
Last Page
8284
Publication Date
1-1-2023
Recommended Citation
A. Kabra et al., "Multi-lingual and Multi-cultural Figurative Language Understanding," Proceedings of the Annual Meeting of the Association for Computational Linguistics, pp. 8269 - 8284, Jan 2023.