Principles from Clinical Research for NLP Model Generalization

Aparna Elangovan, Jiayuan He, Yuan Li, Karin Verspoor


Abstract
The NLP community typically relies on performance of a model on a held-out test set to assess generalization. Performance drops observed in datasets outside of official test sets are generally attributed to “out-of-distribution” effects. Here, we explore the foundations of generalizability and study the factors that affect it, articulating lessons from clinical studies. In clinical research, generalizability is an act of reasoning that depends on (a) *internal validity* of experiments to ensure controlled measurement of cause and effect, and (b) *external validity* or transportability of the results to the wider population. We demonstrate how learning spurious correlations, such as the distance between entities in relation extraction tasks, can affect a model’s internal validity and in turn adversely impact generalization. We, therefore, present the need to ensure internal validity when building machine learning models in NLP. Our recommendations also apply to generative large language models, as they are known to be sensitive to even minor semantic preserving alterations. We also propose adapting the idea of *matching* in randomized controlled trials and observational studies to NLP evaluation to measure causation.
Anthology ID:
2024.naacl-long.127
Volume:
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Kevin Duh, Helena Gomez, Steven Bethard
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2293–2309
Language:
URL:
https://aclanthology.org/2024.naacl-long.127
DOI:
Bibkey:
Cite (ACL):
Aparna Elangovan, Jiayuan He, Yuan Li, and Karin Verspoor. 2024. Principles from Clinical Research for NLP Model Generalization. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 2293–2309, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
Principles from Clinical Research for NLP Model Generalization (Elangovan et al., NAACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.naacl-long.127.pdf
Copyright:
 2024.naacl-long.127.copyright.pdf