A new autonomous system called Empirical Research Assistance (ERA) leverages large language models (LLMs) to generate and optimize disease forecasting software. In a prospective evaluation during the 2025-26 respiratory season, ERA created diverse models for influenza, COVID-19, and RSV. An ensemble of these automatically generated models outperformed the CDC's gold-standard forecasts for all three pathogens and ranked first in official post-season evaluations. This framework offers a way to rapidly scale high-quality disease forecasting, even in data-scarce situations. AI
IMPACT This system demonstrates a path toward rapidly scaling high-quality disease forecasting, potentially improving public health preparedness.
RANK_REASON The item is a research paper detailing a novel system for disease forecasting using LLMs. [lever_c_demoted from research: ic=1 ai=1.0]
- COVID-19
- Empirical Research Assistance (ERA)
- influenza
- large language model
- Sarah Martinson
- United States Centers for Disease Control and Prevention
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
