Structural Ambiguity and its Disambiguation in Language Model Based Parsers: the Case of Dutch Clause Relativization

Gijs Wijnholds, Michael Moortgat


Abstract
This paper addresses structural ambiguity in Dutch relative clauses. By investigating the task of disambiguation by grounding, we study how the presence of a prior sentence can resolve relative clause ambiguities. We apply this method to two parsing architectures in an attempt to demystify the parsing and language model components of two present-day neural parsers. Results show that a neurosymbolic parser, based on proof nets, is more open to data bias correction than an approach based on universal dependencies, although both set-ups suffer from a comparable initial data bias.
Anthology ID:
2023.conll-1.11
Volume:
Proceedings of the 27th Conference on Computational Natural Language Learning (CoNLL)
Month:
December
Year:
2023
Address:
Singapore
Editors:
Jing Jiang, David Reitter, Shumin Deng
Venue:
CoNLL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
155–164
Language:
URL:
https://aclanthology.org/2023.conll-1.11
DOI:
10.18653/v1/2023.conll-1.11
Bibkey:
Cite (ACL):
Gijs Wijnholds and Michael Moortgat. 2023. Structural Ambiguity and its Disambiguation in Language Model Based Parsers: the Case of Dutch Clause Relativization. In Proceedings of the 27th Conference on Computational Natural Language Learning (CoNLL), pages 155–164, Singapore. Association for Computational Linguistics.
Cite (Informal):
Structural Ambiguity and its Disambiguation in Language Model Based Parsers: the Case of Dutch Clause Relativization (Wijnholds & Moortgat, CoNLL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.conll-1.11.pdf