Back to Search Start Over

How Does Data Corruption Affect Natural Language Understanding Models? A Study on GLUE datasets

How Does Data Corruption Affect Natural Language Understanding Models? A Study on GLUE datasets

Authors :
Talman, Aarne
Apidianaki, Marianna
Chatzikyriakidis, Stergios
Tiedemann, Jörg
Publication Year :
2022

Abstract

A central question in natural language understanding (NLU) research is whether high performance demonstrates the models' strong reasoning capabilities. We present an extensive series of controlled experiments where pre-trained language models are exposed to data that have undergone specific corruption transformations. These involve removing instances of specific word classes and often lead to non-sensical sentences. Our results show that performance remains high on most GLUE tasks when the models are fine-tuned or tested on corrupted data, suggesting that they leverage other cues for prediction even in non-sensical contexts. Our proposed data transformations can be used to assess the extent to which a specific dataset constitutes a proper testbed for evaluating models' language understanding capabilities.<br />Comment: *SEM 2022 camera ready version

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2201.04467
Document Type :
Working Paper