AI-Generated Clinical Score Sheets Using Advanced LLMs: Study 2 Raw Data

This dataset contains the raw outputs and source files associated with Study 2 of a project investigating the application of advanced Large Language Models (LLMs) to generate clinical score sheets for animal welfare assessment. Building upon prior research (Berce et al., 2024), this study evaluates the performance of newer LLMs including GPT-4, Claude, and Gemini across multiple animal models and clinical scenarios. The files include prompt templates, LLM responses, scoring breakdowns, and annotations. This release supports transparency and reproducibility in the development of AI-assisted veterinary research tools.本数据集包含某科研项目第二项研究的原始输出与关联源文件，该项目聚焦先进大语言模型（Large Language Models, LLMs）在生成动物福利评估临床评分表领域的应用。本研究基于已有研究成果（Berce等人，2024），对包括GPT-4、Claude与Gemini在内的新一代大语言模型，在多种动物模型与临床场景下的性能表现开展评估。本次发布的文件涵盖提示词模板、大语言模型输出结果、评分细目与标注数据。本数据集的发布旨在助力AI辅助兽医研究工具开发过程的透明度与可复现性提升。