What’s the Meaning of Superhuman Performance in Today’s NLU?
Simone Tedeschi, Johan Bos, Thierry Declerck, Jan Hajič, Daniel Hershcovich, Eduard Hovy, Alexander Koller, Simon Krek
Abstract
In the last five years, there has been a significant focus in Natural Language Processing (NLP) on developing larger Pretrained Language Models (PLMs) and introducing benchmarks such as SuperGLUE and SQuAD to measure their abilities in language understanding, reasoning, and reading comprehension. These PLMs have achieved impressive results on these benchmarks, even surpassing human performance in some cases. This has led to claims of superhuman capabilities and the provocative idea that certain tasks have been solved. In this position paper, we take a critical look at these claims and ask whether PLMs truly have superhuman abilities and what the current benchmarks are really evaluating. We show that these benchmarks have serious limitations affecting the comparison between humans and PLMs and provide recommendations for fairer and more transparent benchmarks.
BibTeX
@inproceedings{tedeschi-etal-2023-whats,
title = "What`s the Meaning of Superhuman Performance in Today`s {NLU}?",
author = "Tedeschi, Simone and
Bos, Johan and
Declerck, Thierry and
Haji{\v{c}}, Jan and
Hershcovich, Daniel and
Hovy, Eduard and
Koller, Alexander and
Krek, Simon and
Schockaert, Steven and
Sennrich, Rico and
Shutova, Ekaterina and
Navigli, Roberto",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.697/",
doi = "10.18653/v1/2023.acl-long.697",
pages = "12471--12491"
}