Automatic Analysis of Flaws in Pre-Trained NLP Models

Richard Eckart de Castilho


Abstract
Most tools for natural language processing today are based on machine learning and come with pre-trained models. In addition, third-parties provide pre-trained models for popular NLP tools. The predictive power and accuracy of these tools depends on the quality of these models. Downstream researchers often base their results on pre-trained models instead of training their own. Consequently, pre-trained models are an essential resource to our community. However, to be best of our knowledge, no systematic study of pre-trained models has been conducted so far. This paper reports on the analysis of 274 pre-models for six NLP tools and four potential causes of problems: encoding, tokenization, normalization and change over time. The analysis is implemented in the open source tool Model Investigator. Our work 1) allows model consumers to better assess whether a model is suitable for their task, 2) enables tool and model creators to sanity-check their models before distributing them, and 3) enables improvements in tool interoperability by performing automatic adjustments of normalization or other pre-processing based on the models used.
Anthology ID:
W16-5203
Volume:
Proceedings of the Third International Workshop on Worldwide Language Service Infrastructure and Second Workshop on Open Infrastructures and Analysis Frameworks for Human Language Technologies (WLSI/OIAF4HLT2016)
Month:
December
Year:
2016
Address:
Osaka, Japan
Editors:
Yohei Murakami, Donghui Lin, Nancy Ide, James Pustejovsky
Venue:
OIAF4HLT
SIG:
Publisher:
The COLING 2016 Organizing Committee
Note:
Pages:
19–27
Language:
URL:
https://aclanthology.org/W16-5203
DOI:
Bibkey:
Cite (ACL):
Richard Eckart de Castilho. 2016. Automatic Analysis of Flaws in Pre-Trained NLP Models. In Proceedings of the Third International Workshop on Worldwide Language Service Infrastructure and Second Workshop on Open Infrastructures and Analysis Frameworks for Human Language Technologies (WLSI/OIAF4HLT2016), pages 19–27, Osaka, Japan. The COLING 2016 Organizing Committee.
Cite (Informal):
Automatic Analysis of Flaws in Pre-Trained NLP Models (Eckart de Castilho, OIAF4HLT 2016)
Copy Citation:
PDF:
https://aclanthology.org/W16-5203.pdf
Code
 UKPLab/coling2016-modelinspector