TWIG (Time‑Causal Wavelet Operator for Irregular Graphs) is a graph‑native neural operator designed for autoregressive surrogate modeling on static irregular graphs. It transforms each node’s history into causal multiscale temporal features, separating recent changes from slower memory components, and propagates these through graph‑wavelet operator blocks with gated pointwise channel mixing. The architecture is causal by construction, enabling closed‑loop forecasting where predictions are recursively reused as future inputs, and it consistently outperforms non‑time‑causal baselines across three irregular‑domain forecasting problems.
arXiv:2606. 19138v1 Announce Type: new Abstract: Neural Controlled Differential Equations (NCDE) provide a powerful continuous-time framework for forecasting time series, but standard graph-based extensions typically learn spatial structure purely from data, even in settings where a directed graph structure is known a priori.
arXiv:2608.27883v1 Announce Type: new Abstract: Physical systems are often modeled by solution operators that map input fields, parameters, geometries, or past states to steady or future physical sta...
The paper introduces a physics‑refined framework for spatiotemporal forecasting on open‑boundary hydrologic graphs, addressing instability caused by missing external boundary forcing. It learns ghost node proxies to approximate unobserved inputs and applies two physics refiners: one enforcing local consistency with two‑hop neighbors, and another using a physics‑guided graph neural operator to reduce long‑horizon drift. Experiments on two real‑world hydrologic graphs show improved prediction accuracy and stability compared to existing learning‑based and physics‑informed models.
arXiv:2606. 11162v1 Announce Type: new Abstract: In this work, we present COGENT, a continuous graph emulator with Neural Ordinary Differential Equations for long-term physical forecasting on irregular geospatial meshes.
arXiv:2606. 03121v1 Announce Type: new Abstract: Multivariate time series forecasting plays a critical role in real-world applications, including weather prediction, stock analysis, and health monitoring.
arXiv:2609.24609v1 Announce Type: new Abstract: Electricity forecasting often involves spatially related signals observed over regions, substations, and feeders, and Graph Neural Networks (GNNs) prov...
The paper introduces m-WCN, an end‑to‑end deep learning framework that neuralizes multi‑wavelet decomposition to jointly extract temporal patterns and frequency components from time series. It enforces orthogonality constraints to produce interpretable multi‑resolution representations, and builds two task‑specific architectures—TFBC for classification and FTB for forecasting—on top of this foundation. Experiments on 64 UCR datasets and seven forecasting benchmarks show that TFBC and FTB outperform baseline models, achieving average improvements of about 20% in both classification and forecasting tasks.
arXiv:2606. 26487v1 Announce Type: cross Abstract: Large language models (LLMs) are attractive for context-aware time series forecasting because they can integrate heterogeneous textual signals, yet their discrete, language-oriented tokenization and embedding interfaces are misaligned with continuous numerical values, often harming numerical ordering and forecasting reliability.
arXiv:2606. 13119v2 Announce Type: replace-cross Abstract: Spatio-Temporal forecasting is crucial in diverse fields, such as transportation, climate, and energy.
arXiv:2606. 13119v1 Announce Type: cross Abstract: Spatio-Temporal forecasting is crucial in diverse fields, such as transportation, climate, and energy.
The paper introduces m-WCN, an end‑to‑end deep learning framework that neuralizes multi‑wavelet decomposition to jointly extract temporal patterns and frequency components from time series. Two task‑specific architectures built on m‑WCN—TFBC for classification and FTB for forecasting—are shown to outperform baseline models on 64 UCR datasets and seven forecasting benchmarks, achieving average improvements of nearly 20% in both tasks. The approach leverages trainable convolutional operators and orthogonality constraints to produce interpretable multi‑resolution representations.
arXiv:2510. 09484v3 Announce Type: replace Abstract: Limited-Area Models (LAMs) enable weather forecasting over regional domains at higher resolutions than what is computationally feasible for global models.