世界银行：2025年平衡创新与严谨：人工智能评估的深思熟虑整合指南报告（指导说明）（英文版）

Within the evolving landscape of artificial intelligence, large language models (LLMs), a type of generative artificial intelligence, offer significant potential for improving the collection, processing, and analysis of large volumes of text data in evaluation. In this note, we present key lessons and good practices for leveraging LLMs based on our recent experiments. The experiments’ results reveal that the LLMs tested could perform text classification quite well, achieving satisfactory reca