POLITICS: Pretraining with Same-story Article Comparison for Ideology Prediction and Stance Detection

Anonymous

Paper Link: https://openreview.net/forum?id=qNDQc26ZxdP

Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)

Abstract: Ideology is at the core of political science research. Yet, there still does not exist general-purpose tools to characterize and predict ideology across different genres of text. To this end, we study Pretrained Language Models using novel ideology-driven pretraining objectives that rely on the comparison of articles on the same story written by media of different ideologies. We further collect a large-scale dataset, consisting of more than 3.6M political news articles, for pretraining. Our model POLITICS outperforms strong baselines and the previous state-of-the-art models on ideology prediction and stance detection tasks. Further analyses show that POLITICS is especially good at understanding long or formally written texts, and is also robust in few-shot learning scenarios.

Response To Ethics Reviews (for Conditionally Accepted Papers Only): Following ethics reviews, we have updated our Ethical Considerations section accordingly. Specifically, for the concern that our tool can "flag" people with unwanted political leaning, we have added the following statements: "Moreover, POLITICS might also be misused to label people with a specific political leaning that they do not want to be associated with. We suggest that when in use the tools should be accompanied with descriptions about their limitations and imperfect performance, as well as allow users to opt out from being the subjects of measurement."

Presentation Mode: This paper will be presented in person in Seattle

Copyright Consent Signature (type Name Or NA If Not Transferrable): Xinliang Frederick Zhang

Copyright Consent Name And Address: University of Michigan, 2260 Hayward St, Ann Arbor, MI 48109