Gender Bias in Contextualized Word Embeddings

04/05/2019
by   Jieyu Zhao, et al.
0

In this paper, we quantify, analyze and mitigate gender bias exhibited in ELMo's contextualized word vectors. First, we conduct several intrinsic analyses and find that (1) training data for ELMo contains significantly more male than female entities, (2) the trained ELMo embeddings systematically encode gender information and (3) ELMo unequally encodes gender information about male and female entities. Then, we show that a state-of-the-art coreference system that depends on ELMo inherits its bias and demonstrates significant bias on the WinoBias probing corpus. Finally, we explore two methods to mitigate such gender bias and show that the bias demonstrated on WinoBias can be eliminated.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/10/2018

Is there Gender bias and stereotype in Portuguese Word Embeddings?

In this work, we propose an analysis of the presence of gender bias asso...
research
06/20/2020

MDR Cluster-Debias: A Nonlinear WordEmbedding Debiasing Pipeline

Existing methods for debiasing word embeddings often do so only superfic...
research
10/06/2020

Robustness and Reliability of Gender Bias Assessment in WordEmbeddings: The Role of Base Pairs

It has been shown that word embeddings can exhibit gender bias, and vari...
research
09/02/2019

It's All in the Name: Mitigating Gender Bias with Name-Based Counterfactual Data Substitution

This paper treats gender bias latent in word embeddings. Previous mitiga...
research
01/23/2019

Attenuating Bias in Word Vectors

Word vector representations are well developed tools for various NLP and...
research
10/20/2022

Choose Your Lenses: Flaws in Gender Bias Evaluation

Considerable efforts to measure and mitigate gender bias in recent years...
research
10/16/2018

Gender Bias in Nobel Prizes

Strikingly few Nobel laureates within medicine, natural and social scien...

Please sign up or login with your details

Forgot password? Click here to reset