Abstract
The meaning of polysemous words often varies in a highly productive yet predictable way. Generalizing the regularity between conventional senses to derive novel word meaning is crucial for automated processing of non-literal language uses such as figurative expressions. We introduce a novel task called systematic word meta-sense extension (SWORME) to test and improve language models’ ability to extend word meaning to denote new semantic domains (also called meta-senses) that bear regular semantic relations with existing senses. We found that language models prefer incremental lexical semantic change toward conceptually similar meta-senses such as logical metonymy, and are much worse at predicting highly non-literal meaning extensions such as metaphors. We propose a novel analogy-based method of word meaning extension, and show that it effectively improves language model systematicity in making both gradual and radical types of meta-sense extension. We further demonstrate that learning systematic meta-sense extensions benefits language models on multiple benchmarks of figurative language understanding.- Anthology ID:
- 2023.emnlp-main.675
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 10953–10966
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.675
- DOI:
- 10.18653/v1/2023.emnlp-main.675
- Cite (ACL):
- Lei Yu. 2023. Systematic word meta-sense extension. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 10953–10966, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Systematic word meta-sense extension (Yu, EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/dois-2013-emnlp/2023.emnlp-main.675.pdf