Neurosymbolic AI as an antithesis to scaling laws.
作為擴展法則對立面的Neurosymbolic AI
PNAS Nexus 2025-05-21
近年來,AI發展越來越仰賴大量資料和高運算資源,導致開發成本高、取得不易,也不太環保。Neurosymbolic AI 結合多種方法,用較少資料和資源,就能達到強大效果,是更永續的選擇。
相關文章PubMedDOI推理
Evaluating large language models in echocardiography reporting: opportunities and challenges.
在心臟超音波報告中評估大型語言模型:機會與挑戰
Eur Heart J Digit Health 2025-05-21
研究團隊用開源大型語言模型微調,讓它們能自動整理心臟超音波報告。其中 EchoGPT(基於 Llama-2)表現最好,產出的總結精簡度幾乎和心臟科醫師寫的一樣好。這技術有望加快報告產出速度又不影響品質,但目前評估 AI 報告的方法還需要改進。
相關文章PubMedDOI推理
Modern artificial intelligence and large language models in graduate medical education: a scoping review of attitudes, applications & practice.
現代人工智慧與大型語言模型於研究所醫學教育中的應用:態度、實務與應用之範疇性回顧
BMC Med Educ 2025-05-20
A framework for evaluating the chemical knowledge and reasoning abilities of large language models against the expertise of chemists.
評估大型語言模型化學知識與推理能力相較於化學專家專業的評估框架
Nat Chem 2025-05-20
Hidden challenges in evaluating spillover risk of zoonotic viruses using machine learning models.
使用機器學習模型評估人畜共通病毒溢出風險的隱藏挑戰
Commun Med (Lond) 2025-05-20
Evaluating the Agreement Between ChatGPT-4 and Validated Mental Health Scales in Older Adults: A Cross-Sectional Study.
ChatGPT-4 與經驗證心理健康量表於年長者間一致性的評估:一項橫斷面研究
Am J Geriatr Psychiatry 2025-05-20
Artificial intelligence for medication-related osteonecrosis of the jaw: a scoping review.
人工智慧於藥物相關性顎骨壞死(medication-related osteonecrosis of the jaw, MRONJ)的應用:一項範疇性回顧
Oral Surg Oral Med Oral Pathol Oral Radiol 2025-05-20
Application of Generative AI to enhance obstetrics and gynecology research.
應用生成式 AI 以提升婦產科研究
Am J Perinatol 2025-05-20
生成式 AI 工具如 ChatGPT、Claude、Gemini,正快速改變婦產科研究,能加速假說產生、文獻回顧、資料分析與論文撰寫,也有助於教育及未來臨床決策。但風險包括產生錯誤、假文獻、隱私疑慮及過度依賴。建議加強人工監督、查核資料、保護隱私並記錄提示。只要妥善管理,AI 可提升研究效率與品質,但仍需持續監控其準確性與影響。
相關文章PubMedDOI推理
Artificial Intelligence Chatbots as Sources of Implant Dentistry Information for the Public: Validity and Reliability Assessment.
人工智慧聊天機器人作為公眾植牙資訊來源之效度與信賴度評估
Eur J Dent 2025-05-20
Comparison of ChatGPT and Internet Research for Clinical Research and Decision-Making in Occupational Medicine: Randomized Controlled Trial.
ChatGPT 與網路搜尋在職業醫學臨床研究與決策中的比較:隨機對照試驗
JMIR Form Res 2025-05-20