Mantis: A Simulation-Grounded Foundation Model for Disease Forecasting
This provides a general-purpose, accurate forecasting tool for public health and epidemiology, especially in settings with limited historical data, though it is incremental as it builds on existing simulation-based approaches.
The paper tackles the problem of infectious disease forecasting in novel outbreaks or low-resource settings by introducing Mantis, a foundation model trained on mechanistic simulations, which achieved lower mean absolute error than all models in the CDC's COVID-19 Forecast Hub and consistently ranked in the top two models across six diseases with diverse transmission modes.
Infectious disease forecasting in novel outbreaks or low-resource settings is hampered by the need for disease-specific data, bespoke training, and expert tuning. We introduce Mantis, a foundation model trained entirely on mechanistic simulations, which enables out-of-the-box forecasting across diseases, regions, and outcomes, even in settings with limited historical data. We evaluated Mantis against 48 forecasting models across six diseases with diverse transmission modes, assessing both point forecast accuracy (mean absolute error) and probabilistic performance (weighted interval score and coverage). Despite using no real-world data during training, Mantis achieved lower mean absolute error than all models in the CDC's COVID-19 Forecast Hub when backtested on early pandemic forecasts. Across all other diseases tested, including respiratory, vector-borne, and waterborne pathogens, Mantis consistently ranked in the top two models across all evaluation metrics. Notably, Mantis generalized to diseases with transmission mechanisms not represented in its training data, demonstrating that it captures fundamental contagion dynamics rather than memorizing disease-specific patterns. These capabilities position Mantis as a practical foundation for disease forecasting: general-purpose, accurate, and deployable where traditional models fail.