表1_五种免费大型语言模型在牙外伤中的表现：一项30天纵向基准研究.docx

数据集获取（Figshare）

• 源站标题：Table 1_Performance of five free large language models in dental trauma: a 30-day longitudinal benchmark study.docx
• DOI：10.3389/froh.2025.1737114.s001
• 发布日期：2025-12-15
• 文件数：1 个
• 文件总体积：724.92 KB
• 源站页面：https://frontiersin.figshare.com/articles/dataset/Table_1_Performance_of_five_free_large_language_models_in_dental_trauma_a_30-day_longitudinal_benchmark_study_docx/30882206

命令行下载

curl -L -o "Table 1_Performance of five free large language models in dental trauma: a 30-day longitudinal benchmark study.docx" "https://ndownloader.figshare.com/files/60365378"

文件清单

• Table 1_Performance of five free large language models in dental trauma: a 30-day longitudinal benchmark study.docx（724.92 KB）

文件总体积 724.92 KB（源站 API 实测）。

数据加载示例（图像类）

from PIL import Image
import glob, os

files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)

# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path)  # 要求 子目录=类别

目录组织与标注格式以源站说明和下载后实际文件为准。

数据说明

ObjectiveTo compare the accuracy and consistency of five large language models (LLMs) in generating responses about dental trauma.
Materials and methodsSixty dichotomous (true/false) questions were submitted daily to each LLM (ChatGPT, Google Gemini, Microsoft Copilot, DeepSeek, and Meta AI) for 30 days, totaling 18,000 responses. All interactions were performed under two prompting conditions (zero-shot and zero-shot with context). LLM responses were compared against the International Association of Dental Traumatology (IADT) guidelines. Statistical analysis was conducted using a generalized linear mixed model (GLMM) with a binomial distribution (α = 0.05), alongside calculation of sensitivity, specificity, accuracy, and area under the ROC curve (AUC) based on the 60-item set. Temporal stability was assessed using the intraclass correlation coefficient ICC.
ResultsAll LLMs achieved accuracy above 85%, with Microsoft Copilot (91.1%) and DeepSeek (90%) performing best; no significant difference was observed between them (p > 0.05), but both outperformed the other models (p  0.90).
ConclusionAll evaluated LLMs, particularly Copilot and DeepSeek, demonstrated high accuracy in providing information on dental trauma, with stable performance over time. While the use of a context prompt did not significantly affect accuracy or stability.

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.3389/froh.2025.1737114.s001（source: https://frontiersin.figshare.com/articles/dataset/Table_1_Performance_of_five_free_large_language_models_in_dental_trauma_a_30-day_longitudinal_benchmark_study_docx/30882206）。