PA · 2026
When AI models look at a political campaign photo, do they make assumptions about the candidate's ideology based on race or gender — even when the ad itself is completely neutral? …
Working Paper · 2026
Examines how vision-language models respond to visual cues of gender and race, documenting systematic associations between demographic features in images and stereotyped outputs. …
Working Paper · 2026
Demonstrates how multilingual language models enable cross-lingual text analysis without translation. Fine-tuning a multilingual model on just one language matches the performance …
FAccT · 2024
Psychologists have long known that people tend to perceive minority and marginalized groups as more uniform and interchangeable than dominant groups — a phenomenon called …
PNAS Nexus · 2024
Uses computational analysis of large language corpora to uncover racial hierarchies embedded in American English. The study reveals systematic associations between race, perceived …
R package for crowdsourced validation of topic models via Amazon Mechanical Turk. Creates and manages tasks that assess both topic coherence and label accuracy, integrating …
PA · 2022
Topic models have become one of the most popular tools for analyzing political text at scale — but are researchers actually checking whether the topics they find measure what they …
APSR · 2017
Content-coding political texts is one of the most labor-intensive tasks in the discipline — but what if you could reliably outsource thousands of judgments to online workers …
