Striking a Balance: Alleviating Inconsistency in Pre-trained Models for Symmetric Classification Tasks

Ashutosh Kumar, Aditya Joshi


Abstract
While fine-tuning pre-trained models for downstream classification is the conventional paradigm in NLP, often task-specific nuances may not get captured in the resultant models. Specifically, for tasks that take two inputs and require the output to be invariant of the order of the inputs, inconsistency is often observed in the predicted labels or confidence scores. We highlight this model shortcoming and apply a consistency loss function to alleviate inconsistency in symmetric classification. Our results show an improved consistency in predictions for three paraphrase detection datasets without a significant drop in the accuracy scores. We examine the classification performance of six datasets (both symmetric and non-symmetric) to showcase the strengths and limitations of our approach.
Anthology ID:
2022.findings-acl.148
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1887–1895
Language:
URL:
https://aclanthology.org/2022.findings-acl.148
DOI:
10.18653/v1/2022.findings-acl.148
Bibkey:
Cite (ACL):
Ashutosh Kumar and Aditya Joshi. 2022. Striking a Balance: Alleviating Inconsistency in Pre-trained Models for Symmetric Classification Tasks. In Findings of the Association for Computational Linguistics: ACL 2022, pages 1887–1895, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Striking a Balance: Alleviating Inconsistency in Pre-trained Models for Symmetric Classification Tasks (Kumar & Joshi, Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-acl.148.pdf
Software:
 2022.findings-acl.148.software.zip
Data
GLUEMRPCPAWSQNLISSTSST-2