Word Embeddings (Also) Encode Human Personality Stereotypes

By Oshin Agarwal, F. Durupinar, N. Badler, Ani Nenkova. 2019

19 citations recorded by the source corpus.

Word representations trained on text reproduce human implicit bias related to gender, race and age. Methods have been developed to remove such bias. Here, we present results that show that human stereotypes exist even for much more nuanced judgments such as personality, for a variety of person identities beyond the typically legally protected attributes and that these are similarly captured in word representations. Specifically, we collected human judgments about a person’s Big Five personality traits formed solely from information about the occupation, nationality or a common noun description of a hypothetical person. Analysis of the data reveals a large number of statistically significant stereotypes in people. We then demonstrate the bias captured in lexical representations is statistically significantly correlated with the documented human bias. Our results, showing bias for a large set of person descriptors for such nuanced traits put in doubt the feasibility of broadly and fairly applying debiasing methods and call for the development of new methods for auditing language technology systems and resources.

Subfields: Artificial Intelligence, Clinical Psychology, Sociology and Political Science, Representation & Self-Supervised Learning, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Clinical Psychology

Hal Daumé Iii

In Sociology and Political Science

In Representation & Self-Supervised Learning

In Safety, Robustness & Interpretability

Citations in the corpus

Language (Technology) is Power: A Critical Survey of “Bias” in NLP (2020, background): “…(2017); McCurdy and Serbetçi (2017); Díaz et al. (2018); Santana et al. (2018); Sutton et al. (2018); Zhang et al. (2018); Zhao et al. (2018a,b); Agarwal et al. (2019); Basta et al. (2019); Bhaskaran and Bhallamudi (2019); Brunet et al. (2019); Cao and Daumé (2019); Chaloner and Maldonado…”
Five sources of bias in natural language processing (2021, background): “…by demographic biases, such as unequal performance for different user groups (Tatman, 2017), mis-identification of speakers and their needs (Criado Perez, 2019) or the proliferation of harmful stereotypes (e.g., Agarwal et al., 2019; Koolen & van Cranenburgh, 2017; Kiritchenko & Moham-mad, 2018).”
Neutralizing Gender Bias in Word Embedding with Latent Disentanglement and Counterfactual Generation (2020, background): “ches have disclosed that the word embeddings contain unexpected biases in their geometry on the embedding space (Bolukbasi et al., 2016; Garg et al., 2018; Ethayarajh et al., 2019; Zhao et al., 2019; Agarwal et al., 2019). The biases reﬂect unwanted stereotypes such as the correlation between the ge”