提出融合多专家的半监督模型，提升小流域十年级日径流预测精度。
HydroFusion-LMF: Semi-Supervised Multi-Network Fusion with Large-Model Adaptation for Long-Term Daily Runoff Forecasting
- 通过可学习分解降低非平稳性，用多类型专家处理残差信号。
- 在10年数据上实现MSE 1.0128、MAE 0.5818，优于最强基线10%以上。
- 适合需要少标注、高适应性的水文预测场景，兼顾可解释性与性能。
小流域十年级日径流预测因趋势漂移、多尺度季节周期、状态突变和稀疏极端事件交织而困难。现有深度模型（如DLinear、TimesNet、PatchTST、TiDE、Nonstationary Transformer、LSTNet、LSTM）通常仅关注单一特征，且未充分利用无标签时段，限制了对状态变化的适应能力。本文提出HydroFusion-LMF框架：（i）采用可学习的趋势-季节-残差分解以降低非平稳性；（ii）将残差输入一组紧凑异构专家（线性精修、频率核、局部Transformer、循环记忆、动态归一化注意力）；（iii）通过依赖水文上下文（如日期相位、前期降水、局部方差、洪水指标及流域属性）的门控机制融合专家输出；（iv）引入半监督多任务目标（复合均方误差/平均绝对误差+极端强调+纳什效率系数/Kling-Gupta效率系数、掩码重建、多尺度对比对齐、增强一致性、方差过滤伪标签）。可选适配器/LoRA层高效注入冻结的时间序列基础编码器。在约10年日尺度数据集上，模型取得MSE 1.0128、MAE 0.5818，较最强基线（DLinear）提升10.2%/10.3%，平均基线提升24.6%/17.1%。结果表明，该框架在保持可解释性的同时显著提升非平稳条件下的标签高效水文预测能力。
原文摘要 · Abstract (English)
Accurate decade-scale daily runoff forecasting in small watersheds is difficult because signals blend drifting trends, multi-scale seasonal cycles, regime shifts, and sparse extremes. Prior deep models (DLinear, TimesNet, PatchTST, TiDE, Nonstationary Transformer, LSTNet, LSTM) usually target single facets and under-utilize unlabeled spans, limiting regime adaptivity. We propose HydroFusion-LMF, a unified framework that (i) performs a learnable trend-seasonal-residual decomposition to reduce non-stationarity, (ii) routes residuals through a compact heterogeneous expert set (linear refinement, frequency kernel, patch Transformer, recurrent memory, dynamically normalized attention), (iii) fuses expert outputs via a hydrologic context-aware gate conditioned on day-of-year phase, antecedent precipitation, local variance, flood indicators, and static basin attributes, and (iv) augments supervision with a semi-supervised multi-task objective (composite MSE/MAE + extreme emphasis + NSE/KGE, masked reconstruction, multi-scale contrastive alignment, augmentation consistency, variance-filtered pseudo-labeling). Optional adapter / LoRA layers inject a frozen foundation time-series encoder efficiently. On a ~10-year daily dataset HydroFusion-LMF attains MSE 1.0128 / MAE 0.5818, improving the strongest baseline (DLinear) by 10.2% / 10.3% and the mean baseline by 24.6% / 17.1%. We observe simultaneous MSE and MAE reductions relative to baselines. The framework balances interpretability (explicit components, sparse gating) with performance, advancing label-efficient hydrologic forecasting under non-stationarity.
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。