Saved in:
Bibliographic Details
Main Authors: Li, Xinzhe, Liu, Ming, Gao, Shang, Buntine, Wray
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.15261
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912069617451008
author Li, Xinzhe
Liu, Ming
Gao, Shang
Buntine, Wray
author_facet Li, Xinzhe
Liu, Ming
Gao, Shang
Buntine, Wray
contents Adversarial robustness, domain generalization and dataset biases are three active lines of research contributing to out-of-distribution (OOD) evaluation on neural NLP models. However, a comprehensive, integrated discussion of the three research lines is still lacking in the literature. In this survey, we 1) compare the three lines of research under a unifying definition; 2) summarize the data-generating processes and evaluation protocols for each line of research; and 3) emphasize the challenges and opportunities for future work.
format Preprint
id arxiv_https___arxiv_org_abs_2306_15261
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Survey on Out-of-Distribution Evaluation of Neural NLP Models
Li, Xinzhe
Liu, Ming
Gao, Shang
Buntine, Wray
Computation and Language
Adversarial robustness, domain generalization and dataset biases are three active lines of research contributing to out-of-distribution (OOD) evaluation on neural NLP models. However, a comprehensive, integrated discussion of the three research lines is still lacking in the literature. In this survey, we 1) compare the three lines of research under a unifying definition; 2) summarize the data-generating processes and evaluation protocols for each line of research; and 3) emphasize the challenges and opportunities for future work.
title A Survey on Out-of-Distribution Evaluation of Neural NLP Models
topic Computation and Language
url https://arxiv.org/abs/2306.15261