Comparing AI-Driven Hypothesis Generation: Evaluating Large Language Models

This comparison explores two AI-driven approaches to scientific hypothesis generation: one leveraging Large Language Models (LLMs) for biomedical hypothesis formulation and the other utilizing an in-context learning framework (FieldSHIFT) to uncover interdisciplinary research insights. By analyzing their dataset construction, methods and evaluation techniques,本研究对比探讨了两种基于人工智能的科学假说生成方法：一种利用大语言模型（LLMs）进行生物医学假说构建，另一种借助上下文学习框架（FieldSHIFT）挖掘跨学科研究洞见。通过分析它们的数据集构建、方法及评估技术，