From Form to Function: A Constructional NLI Benchmark

Claire Bonial, Taylor Pellegrin, Melissa Torgbi, Harish Tayyar Madabushi


Abstract
We present CoGS-NLI, a Natural Language Inference (NLI) evaluation benchmark testing understanding of English phrasal constructions drawn from the Construction Grammar Schematicity (CoGS) corpus. This dataset of 1,500 NLI triples facilitates assessment of constructional understanding in a downstream inference task. We present an evaluation benchmark based on the performance of two language models, where we vary the number and kinds of examples given in the prompt, with and without chain-of-thought prompting. The best-performing model and prompt combination achieves a strong overall accuracy of .94 when provided in-context learning examples with the target phrasal constructions, whereas providing additional general NLI examples hurts performance. This evidences the value of resources explicitly capturing the semantics of phrasal constructions, while our qualitative analysis suggests caveats in assuming this performance indicates a deep understanding of constructional semantics.
Anthology ID:
2025.cxgsnlp-1.18
Volume:
Proceedings of the Second International Workshop on Construction Grammars and NLP
Month:
September
Year:
2025
Address:
Düsseldorf, Germany
Editors:
Claire Bonial, Melissa Torgbi, Leonie Weissweiler, Austin Blodgett, Katrien Beuls, Paul Van Eecke, Harish Tayyar Madabushi
Venues:
CxGsNLP | WS
SIG:
SIGSEM
Publisher:
Association for Computational Linguistics
Note:
Pages:
172–179
Language:
URL:
https://aclanthology.org/2025.cxgsnlp-1.18/
DOI:
Bibkey:
Cite (ACL):
Claire Bonial, Taylor Pellegrin, Melissa Torgbi, and Harish Tayyar Madabushi. 2025. From Form to Function: A Constructional NLI Benchmark. In Proceedings of the Second International Workshop on Construction Grammars and NLP, pages 172–179, Düsseldorf, Germany. Association for Computational Linguistics.
Cite (Informal):
From Form to Function: A Constructional NLI Benchmark (Bonial et al., CxGsNLP 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.cxgsnlp-1.18.pdf