Framing Bias in a Large Language Model: A Diagnostic Accuracy Study of Prompt Effects on ChatGPT’s Melanoma Classification
收藏Digital Commons Data UniCatt2026-04-25 收录
下载链接：
https://unicatt.digitalcommonsdata.com/datasets/tpcf29wdzj
下载链接
链接失效反馈官方服务：
资源简介：
- Mendeley Supplemental Figure 1. Representative examples of the test. Each image was presented six times under different instructions: a neutral baseline prompt, and five framed prompts.
- Mendeley Supplemental File 1. Detailed Materials and methods (Study design, Model Access and Interaction, Dataset Details, Prompting Procedure, Outcome Measures and Statistical Analysis) of the study
提供机构：
Università Cattolica del Sacro Cuore
