Shiraishi Makoto, Tomioka Yoko, Okazaki Mutsumi
Department of Plastic and Reconstructive Surgery, The University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan.
Aesthetic Plast Surg. 2025 Apr;49(7):2154-2155. doi: 10.1007/s00266-024-04193-w. Epub 2024 Jun 18.
In a recent Letter to the Editor authored by Daungsupawong et al. in Aesthetic Plastic Surgery, titled "ChatGPT and Clinical Questions on the Practical Guideline of Blepharoptosis: Correspondence," the authors emphasized important points regarding the input language differences between input and output references. However, advanced versions, such as GPT-4, have shown marginal differences between English and Chinese inputs, possibly because of the use of larger training data. To address this issue, non-English-language-oriented large language models (LLMs) have been developed. The ability of LLMs to refer to existing references varies, with newer models, such as GPT-4, showing higher reference rates than GPT-3.5. Future research should focus on addressing the current limitations and enhancing the effectiveness of emerging LLMs in providing accurate and informative answers to medical questions across multiple languages.Level of Evidence V This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .
在最近由当苏帕翁等人发表在《美容整形外科学》上的一篇致编辑的信中，标题为“ChatGPT与上睑下垂实用指南的临床问题：通信”，作者强调了关于输入和输出参考文献之间输入语言差异的要点。然而，诸如GPT-4等高级版本在英文和中文输入之间显示出微小差异，这可能是由于使用了更大的训练数据。为了解决这个问题，已经开发了非英语导向的大语言模型（LLMs）。大语言模型参考现有参考文献的能力各不相同，诸如GPT-4等较新的模型显示出比GPT-3.5更高的参考率。未来的研究应专注于解决当前的局限性，并提高新兴大语言模型在跨多种语言为医学问题提供准确且信息丰富的答案方面的有效性。证据水平V 本期刊要求作者为每篇文章指定证据水平。有关这些循证医学评级的完整描述，请参阅目录或在线作者指南www.springer.com/00266 。
