On the Limitations of Dataset Balancing: The Lost Battle Against Spurious Correlations

Roy Schwartz, Gabriel Stanovsky


Abstract
Recent work has shown that deep learning models in NLP are highly sensitive to low-level correlations between simple features and specific output labels, leading to over-fitting and lack of generalization. To mitigate this problem, a common practice is to balance datasets by adding new instances or by filtering out “easy” instances (Sakaguchi et al., 2020), culminating in a recent proposal to eliminate single-word correlations altogether (Gardner et al., 2021). In this opinion paper, we identify that despite these efforts, increasingly-powerful models keep exploiting ever-smaller spurious correlations, and as a result even balancing all single-word features is insufficient for mitigating all of these correlations. In parallel, a truly balanced dataset may be bound to “throw the baby out with the bathwater” and miss important signal encoding common sense and world knowledge. We highlight several alternatives to dataset balancing, focusing on enhancing datasets with richer contexts, allowing models to abstain and interact with users, and turning from large-scale fine-tuning to zero- or few-shot setups.
Anthology ID:
2022.findings-naacl.168
Volume:
Findings of the Association for Computational Linguistics: NAACL 2022
Month:
July
Year:
2022
Address:
Seattle, United States
Editors:
Marine Carpuat, Marie-Catherine de Marneffe, Ivan Vladimir Meza Ruiz
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2182–2194
Language:
URL:
https://aclanthology.org/2022.findings-naacl.168
DOI:
10.18653/v1/2022.findings-naacl.168
Bibkey:
Cite (ACL):
Roy Schwartz and Gabriel Stanovsky. 2022. On the Limitations of Dataset Balancing: The Lost Battle Against Spurious Correlations. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 2182–2194, Seattle, United States. Association for Computational Linguistics.
Cite (Informal):
On the Limitations of Dataset Balancing: The Lost Battle Against Spurious Correlations (Schwartz & Stanovsky, Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-naacl.168.pdf
Video:
 https://aclanthology.org/2022.findings-naacl.168.mp4
Data
DROPHellaSwagReCoRDSWAGVisual Question Answering v2.0WinoGrande