Skip to main navigation menu Skip to main content Skip to site footer

Vol. 9 (2021)

TACL approved

Erratum: Measuring and Improving Consistency in Pretrained Language Models

Published 2022-01-04

  • Yanai Elazar
    • ,
  • Nora Kassner
    • ,
  • Shauli Ravfogel
    • ,
  • Abhilasha Ravichander
    • ,
  • Eduard Hovy
    • ,
  • Hinrich Schütze
    • ,
  • Yoav Goldberg

Yanai Elazar
Bar-Ilan University, AI2

Nora Kassner
LMU

Shauli Ravfogel
Bar-Ilan University, AI2

Abhilasha Ravichander
CMU

Eduard Hovy
CMU

Hinrich Schütze
LMU

Yoav Goldberg
Bar-Ilan University, AI2

Abstract

During production of this paper, an error was introduced to the formula on the bottom of the right column of page 1020. In the last two terms of the formula, the n and m subscripts were swapped. The correct formula is:

Lc=∑kn=1∑km=n+1DKL(Qrin∥Qrim)+DKL(Qrim∥Qrin)

The paper has been updated.

Article at MIT Press