[COLING 2022]: CommunityLM: Probing Partisan Worldviews from Language Models - View it on GitHub