Measuring and Improving Consistency in Pretrained Language Models

MCML Authors

Nora Kassner

* Former Member

→ Group Hinrich Schütze
 Computational Linguistics

Hinrich Schütze

Prof. Dr.

Principal Investigator

Computational Linguistics

Abstract

Consistency of a model—that is, the invariance of its behavior under meaning-preserving alternations in its input—is a highly desirable property in natural language processing. In this paper we study the question: Are Pretrained Language Models (PLMs) consistent with respect to factual knowledge? To this end, we create ParaRel, a high-quality resource of cloze-style query English paraphrases. It contains a total of 328 paraphrases for 38 relations. Using ParaRel, we show that the consistency of all PLMs we experiment with is poor— though with high variance between relations. Our analysis of the representational spaces of PLMs suggests that they have a poor structure and are currently not suitable for representing knowledge robustly. Finally, we propose a method for improving model consistency and experimentally demonstrate its effectiveness.

article

---

Transactions of the Association for Computational Linguistics

9. Dec. 2021.

---

Authors

Y. Elazar • N. Kassner • S. Ravfogel • A. Ravichander • E. Hovy • H. Schütze • Y. Goldberg

---

Links

DOI

---

Research Area

 B2 | Natural Language Processing