Research report: Explaining alignment between LLM classifications of political arguments and human judgments
A study of Swiss referendum debates reveals that large language models align with human evaluations of political arguments through surface ...
