Abstract
In this chapter, we discuss how behavioral patterns in NLP models can be analyzed. We discuss how models can be systematically tested for linguistic knowledge and introduce methods to diagnose cognitively implausible sensitivity against perturbations.
Original language | English |
---|---|
Title of host publication | Synthesis Lectures on Human Language Technologies |
Publisher | Springer Nature |
Pages | 61-88 |
Number of pages | 28 |
DOIs | |
Publication status | Published - 2024 |
Publication series
Name | Synthesis Lectures on Human Language Technologies |
---|---|
Volume | Part F1653 |
ISSN (Print) | 1947-4040 |
ISSN (Electronic) | 1947-4059 |
Bibliographical note
Publisher Copyright:© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.