Are Structural Concepts Universal in Transformer Language Models? Towards Interpretable Cross-Lingual Generalization

Ningyu Xu, Qi Zhang, Jingting Ye, Menghan Zhang, Xuanjing Huang


Abstract
Large language models (LLMs) have exhibited considerable cross-lingual generalization abilities, whereby they implicitly transfer knowledge across languages. However, the transfer is not equally successful for all languages, especially for low-resource ones, which poses an ongoing challenge. It is unclear whether we have reached the limits of implicit cross-lingual generalization and if explicit knowledge transfer is viable. In this paper, we investigate the potential for explicitly aligning conceptual correspondence between languages to enhance cross-lingual generalization. Using the syntactic aspect of language as a testbed, our analyses of 43 languages reveal a high degree of alignability among the spaces of structural concepts within each language for both encoder-only and decoder-only LLMs. We then propose a meta-learning-based method to learn to align conceptual spaces of different languages, which facilitates zero-shot and few-shot generalization in concept classification and also offers insights into the cross-lingual in-context learning phenomenon. Experiments on syntactic analysis tasks show that our approach achieves competitive results with state-of-the-art methods and narrows the performance gap between languages, particularly benefiting those with limited resources.
Anthology ID:
2023.findings-emnlp.931
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
13951–13976
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.931
DOI:
10.18653/v1/2023.findings-emnlp.931
Bibkey:
Cite (ACL):
Ningyu Xu, Qi Zhang, Jingting Ye, Menghan Zhang, and Xuanjing Huang. 2023. Are Structural Concepts Universal in Transformer Language Models? Towards Interpretable Cross-Lingual Generalization. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 13951–13976, Singapore. Association for Computational Linguistics.
Cite (Informal):
Are Structural Concepts Universal in Transformer Language Models? Towards Interpretable Cross-Lingual Generalization (Xu et al., Findings 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/naacl24-info/2023.findings-emnlp.931.pdf