[1911.01485] Assessing Social and Intersectional Biases in Contextualized Word Representationsopen searchopen navigation menucontact arXivsubscribe to arXiv mailings

Social bias in machine learning has drawn significant attention, with work ranging from demonstrations of bias in a multitude of applications, curating definitions of fairness for different contexts, to developing algorithms to mitigate bias. In natural language processing, gender bias has been shown to exist in context-free word embeddings. Recently, contextual word representations have outperformed word embeddings in several downstream NLP tasks. These word representations are conditioned on their context within a sentence, and can also be used to encode the entire sentence. In this paper, we analyze the extent to which state-of-the-art models for contextual word representations, such as BERT and GPT-2, encode biases with respect to gender, race, and intersectional identities. Towards this, we propose assessing bias at the contextual word level. This novel approach captures the contextual effects of bias missing in context-free word embeddings, yet avoids confounding effects that und

1 mentions: @otakumesi
Date: 2020/08/22 11:21

Related Entries

Read more Are Pre-trained Language Models Aware of Phrases? Simple but Strong Baselines for Grammar Induction ...
0 users, 1 mentions 2020/02/10 12:54
Read more [1904.12324] OPIEC: An Open Information Extraction Corpuscontact arXivarXiv Twitter
0 users, 1 mentions 2020/03/20 00:51
Read more nlp — nlp 0.4.0 documentation
0 users, 1 mentions 2020/08/27 00:52
Read more GitHub - ResponsiblyAI/responsibly: Toolkit for Auditing and Mitigating Bias and Fairness of Machine...
0 users, 1 mentions 2020/08/27 05:21
Read more GitHub - rudinger/winogender-schemas: Data for evaluating gender bias in coreference resolution syst...
0 users, 1 mentions 2020/09/08 08:21