The CAP Babel Machine (Phase 2)

Manually labeling large corpora can be tedious and often demands extensive domain expertise. Recent advancements in machine learning and natural language processing hold the potential for language models to surpass human-level accuracy in text classification tasks.

In our experiments, we fine-tuned Large Language Models (LLMs) across various language and domain corpora using 21 categories from the Comparative Agendas Project codebook. This project is an update for the project "The CAP Babel Machine." In Phase 2, we fine-tuned XLM-RoBERTa models for new language-domain combinations to extend the capabilities of our previous pipeline. We also leveraged open neural machine translation models to augment our existing training data.