EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts 事件

EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts arXiv:2606.05513v1 Announce Type: cross Abstract: Epidemic LLM forecasters are usually trained and evaluated as static supervised models, whereas operational pandemic forecasting is a streaming process in which labels arrive after predictions and disease regimes shift over time. We study this mismatch in weekly COVID-19 hospitalization trend forecasting across five variant regimes. We introduce EpiEvolve, a s