Proceedings

FLASH: Flexible Learning of Adaptive Sampling from History in Temporal Graph Neural Networks

Or Feldman, Krishna Sri Ipsit Mantri, Carola-Bibiane Schönlieb, Chaim Baskin, Moshe Eliasof

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 4278-4285.

https://doi.org/10.24963/ijcai.2026/476

Keywords

  • Machine Learning: Deep learning architectures
  • Machine Learning: Geometric learning

Abstract

Aggregating temporal signals from historic interactions is a key step in future link prediction on dynamic graphs. However, incorporating long histories is resource-intensive. Hence, temporal graph neural networks (TGNNs) often rely on historical neighbors sampling heuristics such as uniform sampling or recent neighbors selection. These heuristics are static and fail to adapt to the underlying graph structure. We introduce FLASH, a learnable and graph-adaptive neighborhood selection mechanism that generalizes existing heuristics. FLASH integrates seamlessly into TGNNs and is trained end-to-end using a self-supervised ranking loss. We provide theoretical evidence that commonly used heuristics hinder TGNNs performance, motivating our design. Extensive experiments across multiple benchmarks demonstrate consistent and significant performance improvements for TGNNs equipped with FLASH

Or Feldman, Krishna Sri Ipsit Mantri, Carola-Bibiane Schönlieb, Chaim Baskin, and Moshe Eliasof. 2026. FLASH: Flexible Learning of Adaptive Sampling from History in Temporal Graph Neural Networks. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 4278–4285. https://doi.org/10.24963/ijcai.2026/476
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards