Generative Pre Trained Transformer
Generative Pre-trained Transformers (GPTs) are large language models designed to generate human-like text and increasingly, other data modalities. Current research focuses on improving GPT performance across diverse applications, including code generation, scientific modeling (e.g., seismic velocity modeling, cellular automata simulation), and healthcare (e.g., EHR generation, medical image analysis), often employing techniques like retrieval-augmented generation and in-context learning to enhance capabilities. The ability of GPTs to process and generate various data types makes them a powerful tool with significant implications for numerous fields, driving advancements in both scientific understanding and practical applications.
Papers
Hidden flaws behind expert-level accuracy of multimodal GPT-4 vision in medicine
Qiao Jin, Fangyuan Chen, Yiliang Zhou, Ziyang Xu, Justin M. Cheung, Robert Chen, Ronald M. Summers, Justin F. Rousseau, Peiyun Ni, Marc J Landsman, Sally L. Baxter, Subhi J. Al'Aref, Yijia Li, Alex Chen, Josef A. Brejt, Michael F. Chiang, Yifan Peng, Zhiyong Lu
Enhancing Robustness of LLM-Synthetic Text Detectors for Academic Writing: A Comprehensive Analysis
Zhicheng Dou, Yuchen Guo, Ching-Chun Chang, Huy H. Nguyen, Isao Echizen