Time series are the primary data type used to record dynamic system measurements and
generated in great volume by both physical sensors and online processes (virtual sensors) …

Graph Neural Networks (GNNs) have become the leading paradigm for learning on (static)
graph-structured data. However, many real-world systems are dynamic in nature, since the …

Abstract We present the Temporal Graph Benchmark (TGB), a collection of challenging and
diverse benchmark datasets for realistic, reproducible, and robust evaluation of machine …

Graph condensation, which reduces the size of a large-scale graph by synthesizing a small-
scale condensed graph as its substitution, has immediate benefits for various graph learning …

Graph contrastive learning (GCL) alleviates the heavy reliance on label information for
graph representation learning (GRL) via self-supervised learning schemes. The core idea is …

We propose DyGFormer, a new Transformer-based architecture for dynamic graph learning.
DyGFormer is conceptually simple and only needs to learn from nodes' historical first-hop …

Temporal graphs are widely used to model dynamic systems with time-varying interactions.
In real-world scenarios, the underlying mechanisms of generating future interactions in …

Real-world graphs generally have only one kind of tendency in their connections. These
connections are either homophilic-prone or heterophily-prone. While graphs with homophily …

Knowledge graphs (KGs), as a structured form of knowledge representation, have been
widely applied in the real world. Recently, few-shot knowledge graph completion (FKGC) …

Multivariate time-series anomaly detection is critically important in many applications,
including retail, transportation, power grid, and water treatment plants. Existing approaches …