aiqbench analysis
ChronoSpike addresses the efficiency-expressiveness trade-off in dynamic graph learning by combining spiking neural networks with Transformer encoders to reduce memory complexity while maintaining global context. The model demonstrates significant performance gains over recurrent and attention-based baselines on standard benchmarks, offering a viable path toward scalable event-driven graph analysis. Its theoretical guarantees for gradient stability and bounded membrane potentials provide a robust foundation for deploying these models in real-time systems.
ChronoSpike: An Adaptive Spiking Graph Neural Network for Dynamic Graphs
arXiv:2602.01124v4 Announce Type: replace Abstract: Dynamic graph representation learning requires capturing both structural relations and temporal evolution, yet existing approaches face a core trade-off: attention-based methods offer expressiveness at O(T^2) complexity, while recurrent architectures suffer from gradient pathologies and dense state storage. Spiking neural networks provide event-driven efficiency but are constrained by sequential propagation, binary information loss, and local aggregation that lacks global context. We propose ChronoSpike, an adaptive spiking graph neural network that integrates learnable LIF neurons with per-channel membrane dynamics, multi-head spatially-attentive aggregation over continuous features, and a lightweight Transformer temporal encoder. This design enables fine-grained local modeling and long-range dependency capture with O(T.d) activation/state memory and an additional O(T^2) per-node attention term that remains small for the horizons evaluated here. ChronoSpike outperforms twelve state-of-the-art baselines on three large benchmarks by 2.0%~Macro-F1 and 2.4%~Micro-F1 on average while achieving 3-10 times faster training than recurrent methods with a constant 105K-parameter budget independent of graph size. We provide theoretical guarantees for membrane potential boundedness, gradient flow stability under contraction factor \r{ho}<1, and BIBO stability; interpretability analyses reveal heterogeneous temporal receptive fields and a learned primacy effect with 83-88% sparsity. Our code is publicly available at: https://github.com/Abrar2652/ChronoSpike
Abstract:Dynamic graph representation learning requires capturing both structural relations and temporal evolution, yet existing approaches face a core trade-off: attention-based methods offer expressiveness at O(T^2) complexity, while recurrent architectures suffer from gradient pathologies and dense state storage. Spiking neural networks provide event-driven efficiency but are constrained by sequential propagation, binary information loss, and local aggregation that lacks global context. We propose ChronoSpike, an adaptive spiking graph neural network that integrates learnable LIF neurons with per-channel membrane dynamics, multi-head spatially-attentive aggregation over continuous features, and a lightweight Transformer temporal encoder. This design enables fine-grained local modeling and long-range dependency capture with O(T.d) activation/state memory and an additional O(T^2) per-node attention term that remains small for the horizons evaluated here. ChronoSpike outperforms twelve state-of-the-art baselines on three large benchmarks by 2.0%~Macro-F1 and 2.4%~Micro-F1 on average while achieving 3-10 times faster training than recurrent methods with a constant 105K-parameter budget independent of graph size. We provide theoretical guarantees for membrane potential boundedness, gradient flow stability under contraction factor \r{ho}<1, and BIBO stability; interpretability analyses reveal heterogeneous temporal receptive fields and a learned primacy effect with 83-88% sparsity. Our code is publicly available at: this https URL
Machine Learning (cs.LG)
arXiv:2602.01124 [cs.LG]
(or arXiv:2602.01124v4 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2602.01124 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 285 words
Related coverage
Build an explainable next-best-product recommendation system for banking on AWS
AWS · Jul 24, 2026
Molybdenum disulfide transistors reshape spiking neural networks
Quantum Zeitgeist · Sep 14, 2026
ARES: Anomaly Recognition Model For Edge Streams
arXiv · Jul 31, 2026
Continuous Spiking Graph Neural Networks
arXiv · Sep 22, 2026