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.