OmniST：无需固定传感器数量的时空预测模型，新增站点也能给出带覆盖保证的区间
OmniST: Topology-Invariant Spatiotemporal Forecasting with Guaranteed-Coverage Uncertainty at Unseen Sensor Locations
气象站、空气监测这类传感器网络常增减站点，普通图神经网络直接失效。这篇 arXiv 论文的做法能让模型在没见过的站点也输出预测加置信区间，做时空数据的朋友可以看看。
OmniST 是一种拓扑不变时空预测模型，计算过程不依赖节点数量，训练时的邻接矩阵不再绑定固定传感器数 N。它用坐标 tokenization 加 k 近邻注意力，每次前向传播直接从地理坐标计算，节点时序处理用双向 GRU。在德里空气质量数据（37 站，2019-2023）上，30 天 PM2.5 MAE 为 31.0 ug/m3，随机移除 40% 站点后性能最多下降 1.1%。对未见过的站点做归纳预测时，它比最近邻插值低约 12%，比学习型 kriging GNN（IGNNK）低约 21%，并给出分布无关的覆盖保证预测区间。结合滚动自适应 conformal 校准后，覆盖率从静态方案的后期 0.50 恢复到 0.96。该保证在 PeMS-BAY 全量 325 个真实传感器上依然成立。
OmniST: Topology-Invariant Spatiotemporal Forecasting with Guaranteed-Coverage Uncertainty at Unseen Sensor Locations
Standard spatiotemporal graph convolutional networks need a fixed-size adjacency matrix, coupling model weights to a specific sensor count N; they cannot produce any output at a sensor absent from training. We present OmniST, a topology-invariant forecaster whose computation has zero structural dependency on the number of nodes, and which, composed with inductive split-conformal calibration, gives distribution-free, coverage-guaranteed prediction intervals at arbitrary (including never-seen) sensor locations, preserving that guarantee as sensors are added or removed. Topology invariance follows from coordinate-based tokenization, k-nearest-neighbor attention from geographic coordinates at each forward pass, and per-node temporal processing via a bidirectional GRU. On real Delhi air-quality data (37 stations, 2019-2023), OmniST attains a 30-day PM2.5 MAE of 31.0 ug/m3 and degrades at most 1.1 percent when 40 percent of stations are removed. Its inductive forecasts at unseen stations beat nearest-neighbor interpolation (about 12 percent) and a learned kriging GNN (IGNNK, about 21 percent), and, unlike either, carry a finite-sample coverage guarantee. The guarantee holds at 90 percent or above under 30-50 percent node dropout across four real datasets over three domains (Delhi and Beijing air quality, NOAA GHCN weather, Caltrans PeMSD7-M traffic), and still holds at 325 real sensors (full PeMS-BAY). We also fix a temporal-drift failure: static conformal under-covers over time (about 0.50 late), but a rolling adaptive-conformal estimator restores target coverage (0.89 overall, 0.96 late) at the cost of wider intervals.