What does memory retrieval leave on the table? Modelling the Cost of Semi-Compositionality with MINERVA2 and sBERT

Sydelle De Souza, Ivan Vegner, Francis Mollica, Leonidas A. A. Doumas


Abstract
Despite being ubiquitous in natural language, collocations (e.g., kick+habit) incur a unique processing cost, compared to compositional phrases (kick+door) and idioms (kick+bucket). We confirm this cost with behavioural data as well as MINERVA2, a memory model, suggesting that collocations constitute a distinct linguistic category. While the model fails to fully capture the observed human processing patterns, we find that below a specific item frequency threshold, the model’s retrieval failures align with human reaction times across conditions. This suggests an alternative processing mechanism that activates when memory retrieval fails.
Anthology ID:
2025.conll-1.19
Volume:
Proceedings of the 29th Conference on Computational Natural Language Learning
Month:
July
Year:
2025
Address:
Vienna, Austria
Editors:
Gemma Boleda, Michael Roth
Venues:
CoNLL | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
291–311
Language:
URL:
https://preview.aclanthology.org/landing_page/2025.conll-1.19/
DOI:
Bibkey:
Cite (ACL):
Sydelle De Souza, Ivan Vegner, Francis Mollica, and Leonidas A. A. Doumas. 2025. What does memory retrieval leave on the table? Modelling the Cost of Semi-Compositionality with MINERVA2 and sBERT. In Proceedings of the 29th Conference on Computational Natural Language Learning, pages 291–311, Vienna, Austria. Association for Computational Linguistics.
Cite (Informal):
What does memory retrieval leave on the table? Modelling the Cost of Semi-Compositionality with MINERVA2 and sBERT (De Souza et al., CoNLL 2025)
Copy Citation:
PDF:
https://preview.aclanthology.org/landing_page/2025.conll-1.19.pdf