SparCAssist

A Model Risk Assessment Assistant Based on Sparse Generated Counterfactuals

Conference Paper (2022)
Author(s)

Zijian Zhang (L3S Research Center)

Vinay Setty (L3S Research Center, University of Stavanger)

Avishek Anand (TU Delft - Web Information Systems, L3S Research Center)

Research Group
Web Information Systems
Copyright
© 2022 Zijian Zhang, Vinay Setty, A. Anand
DOI related publication
https://doi.org/10.1145/3477495.3531677
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 Zijian Zhang, Vinay Setty, A. Anand
Research Group
Web Information Systems
Pages (from-to)
3219-3223
ISBN (electronic)
978-1-4503-8732-3
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We introduce SparCAssist, a general-purpose risk assessment tool for the machine learning models trained for language tasks. It evaluates models' risk by inspecting their behavior on counterfactuals, namely out-of-distribution instances generated based on the given data instance. The counterfactuals are generated by replacing tokens in rational subsequences identified by ExPred, while the replacements are retrieved using HotFlip or the Masked-Language-Model-based algorithms. The main purpose of our system is to help the human annotators to assess the model's risk on deployment. The counterfactual instances generated during the assessment are the by-product and can be used to train more robust NLP models in the future.

Files

3477495.3531677.pdf
(pdf | 1.31 Mb)
- Embargo expired in 01-07-2023
License info not available