Applying large language model for automated quality scoring of radiology requisitions using a standardized criteria.
應用大型語言模型,利用標準化評分標準自動評分放射科檢查申請單的品質
Eur Radiol 2025-08-20
SciToolAgent: a knowledge-graph-driven scientific agent for multitool integration.
SciToolAgent:一個基於知識圖譜驅動的多工具整合科學代理系統
Nat Comput Sci 2025-08-20
SciToolAgent 是一款由大型語言模型驅動的 AI 科學助手,能自動操作上百種生物、化學和材料科學工具。它會用知識圖譜來聰明選擇工具,內建安全機制確保合乎倫理。基準測試顯示效能優於現有方法,能自動化複雜科學流程,讓專家和一般人都能輕鬆使用先進研究工具。
相關文章PubMedDOI推理
A novel sequence-based transformer model architecture for integrating multi-omics data in preterm birth risk prediction.
用於早產風險預測之多重組學資料整合的創新序列式 transformer 模型架構
NPJ Digit Med 2025-08-20
Artificial intelligence in clinical data analysis: A review of large language models, foundation models, digital twins, and allergy applications.
臨床資料分析中的人工智慧:大型語言模型、基礎模型、數位分身與過敏應用之綜述
Allergol Int 2025-08-20
Use of artificial intelligence in planning postoperative nursing care in laparoscopic cholecystectomy patients: Comparison of ChatGPT and student practice.
人工智慧於腹腔鏡膽囊切除術後護理計畫中的應用:ChatGPT 與護理學生實務的比較
Nurse Educ Pract 2025-08-20
REaMA: Building Biomedical Relation Extraction Specialized Large Language Models Through Instruction Tuning.
REaMA:透過指令微調建構專門用於生物醫學關係抽取的大型語言模型
IEEE Trans Neural Netw Learn Syst 2025-08-20
Assessment of the Accuracy of Modern Artificial Intelligence Chatbots in Responding to Endodontic Queries.
現代人工智慧聊天機器人在回應牙髓病學相關問題時的準確性評估
Aust Endod J 2025-08-20
這項研究發現,ChatGPT-3.5 在牙髓治療是非題的答題準確率最高(80%),優於 ChatGPT-4、Gemini 1.5 Pro 和 Gemini 1.5 Flash。不過,所有 AI 聊天機器人的答案和專家共識的吻合度都偏低。研究建議,臨床決策時不能只靠 AI,未來還需要更深入的研究來驗證其可靠性。
相關文章PubMedDOI推理