让疫情预测模型自动进化，应对病毒变异带来的变化。
EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts
- 用记忆存储历史预测结果，结合延迟反馈动态调整策略。
- 在疫情变异后恢复速度提升，准确率达0.629，优于基准模型。
- 适合需要持续更新的实时公共卫生预测系统使用。
流行病大语言模型通常作为静态监督模型训练与评估，但实际疫情预测是流式过程：预测先于标签到达，且疾病传播模式随时间发生结构性转变。本文研究了在五个变异株阶段下的周度新冠住院趋势预测问题。提出EpiEvolve，一种自进化代理，其核心为固定权重的预训练大语言模型，并通过构建分层情景记忆，存储预测结果，反思延迟标签，检索当前阶段相关案例，并将重复错误提炼为策略规则。该机制使模型能在后续周次中复用自身历史预测与结果，遵循时间顺序协议避免未来信息泄露。在流式数据集上，EpiEvolve平均准确率达0.629，优于静态基线模型（0.561）和外部CDC集成模型（0.325），并在变异后恢复周期从5周缩短至2周。消融实验表明，反思、策略性记忆和阶段感知检索均对性能提升有贡献。
原文摘要 · Abstract (English)
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 self-evolving agent that wraps an LLM forecaster trained on the warm-start period and keeps its weights fixed during streaming. EpiEvolve adapts by storing forecast outcomes in a hierarchical episodic memory, reflecting on delayed labels, retrieving cases relevant to the current regime, and distilling recurring errors into strategic rules. The resulting context lets the forecaster reuse its own past predictions and outcomes in later weeks while following a chronological protocol that prevents future leakage. On the streaming dataset, EpiEvolve reaches $0.629$ average accuracy, compared with $0.561$ for the static backbone and $0.325$ for the external CDC ensemble, and reduces recovery lag after regime shifts from $5$ to $2$ weeks. Ablations show that reflection, strategic memory, and regime-aware retrieval each contribute to the gains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。