Theory-Grounded Measurement of U.S. Social Stereotypes in English Language Models
By Yang Trista Cao, Anna Sotnikova, Hal Daumé Iii, Rachel Rudinger, L. Zou. 2022
63 citations recorded by the source corpus.
NLP models trained on text have been shown to reproduce human stereotypes, which can magnify harms to marginalized groups when systems are deployed at scale. We adapt the Agency-Belief-Communion (ABC) stereotype model of Koch et al. (2016) from social psychology as a framework for the systematic study and discovery of stereotypic group-trait associations in language models (LMs). We introduce the sensitivity test (SeT) for measuring stereotypical associations from language models. To evaluate SeT and other measures using the ABC model, we collect group-trait judgments from U.S.-based subjects to compare with English LM stereotypes. Finally, we extend this framework to measure LM stereotyping of intersectional identities.
Subfields: Artificial Intelligence, Clinical Psychology, General Social Sciences, Large Language Models, AI Ethics, Fairness & Society
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
- Willem H. Zuidema
- Oskar van der Wal
- Alina Leidinger
- Leendert van Maanen
- Dominik Bachmann
- Katrin Schulz
- Yoav Goldberg
- Yonatan Belinkov
In General Social Sciences
In Large Language Models
Citations in the corpus
- Aligning Language Models with Preferences through f-divergence Minimization (2023, neutral): “2022) or equal treatment (Cao et al., 2022).”
- Undesirable biases in NLP: Averting a crisis of measurement (2022, methodology): “…NLP bias measures against data from human behavior: e.g., common knowledge of stereotypes, results from psychological tests (Caliskan et al., 2017; Cao et al., 2022a), or statistics of the gender division for occupations in the USA (Caliskan et al., 2017; Zhao et al., 2018; Webster et al.,…”