Paper · arXiv 2409.02078
Political DEBATE: Efficient Zero-shot and Few-shot Classifiers for Political Text
Michael Burnham, Kayla Kahn, Ryan Yank Wang, Rachel X. Peng
Political DEBATE models achieve high efficiency and accuracy in zero-shot and few-shot political document classification using minimal training data and are openly sourced.
Abstract
Social scientists quickly adopted large language models due to their ability to annotate documents without supervised training, an ability known as zero-shot learning. However, due to their compute demands, cost, and often proprietary nature, these models are often at odds with replication and open science standards. This paper introduces the Political DEBATE (DeBERTa Algorithm for Textual Entailment) language models for zero-shot and few-shot classification of political documents. These models are not only as good, or better than, state-of-the art large language models at zero and few-shot classification, but are orders of magnitude more efficient and completely open source. By training the models on a simple random sample of 10-25 documents, they can outperform supervised classifiers trained on hundreds or thousands of documents and state-of-the-art generative models with complex, engineered prompts. Additionally, we release the PolNLI dataset used to train these models -- a corpus of over 200,000 political documents with highly accurate labels across over 800 classification tasks.