Researchers at the Icahn School of Medicine at Mount Sinai have unveiled strategies for integrating large language models (LLMs)—a type of artificial intelligence (#AI)—into health systems, balancing cost efficiency with high performance. “Our findings offer a road map for health care systems to use advanced AI tools effectively, potentially reducing API call costs for LLMs by up to 17-fold, while ensuring reliable performance even under heavy workloads,” says co-senior author Girish N. Nadkarni, MD, MPH. “Our goal was to address the challenges of scaling LLMs in health care by identifying practical ways to cut costs without sacrificing functionality. We ‘stress tested’ these models to see how they perform with multiple tasks simultaneously and developed strategies to maintain efficiency and stability,” adds first author Eyal Klang, MD, Director of the Generative AI Research Program in the D3M at Icahn Mount Sinai. Learn more: https://bit.ly/3OptKF6
Co-Founder of Altrosyn and DIrector at CDTECH | Inventor | Manufacturer
3 天前Cost-efficiency in LLMs is crucial for healthcare adoption. Stress testing reveals vulnerabilities under heavy workloads. How do you envision fine-tuning these models for real-time, dynamic patient data streams?