mlsanityData to 14 Oct 2026

ASPECT: Node-Level Adaptive Spectral Fusion for Graph Contrastive Learning

arXiv 2604.01878 ↗First posted 2 Apr 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Spectral graph contrastive learning often constructs low- and high-frequency views to capture complementary graph signals, but these views are commonly combined by graph-level or node-agnostic fusion rules. We show that graph-level fusion can incur irreducible regret on mixed graphs with separated node-wise spectral preferences. Motivated by this result, we propose ASPECT, a spectral graph contrastive learning method that adaptively fuses low- and high-frequency views at the node level. ASPECT learns a node-wise spectral policy and regularizes it using channel-wise contrastive evidence, enabling different nodes to use different spectral mixtures. We further introduce ASPECT-S, an optional stability-aware extension that uses generated graph-structure and feature perturbations to obtain empirical channel-wise sensitivity estimates, together with a Rayleigh-based spectral search bias for producing informative perturbations. Experiments on homophilic and heterophilic benchmarks show that ASPECT improves representation quality over competitive spectral and graph contrastive baselines, while ASPECT-S further improves performance under joint graph-structure and feature perturbations.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 37 of its 44 references that are in the map. First posted on arXiv 2 Apr 2026.

Builds on

37 of 44 references.

References · 37

  1. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  2. 2020Graph Contrastive Learning with Augmentationscited by 1,208
  3. 2018Deep Graph Infomaxcited by 1,106
  4. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 826
  5. 2020Contrastive Multi-View Representation Learning on Graphscited by 652
  6. 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 595
  7. 2019Multi-scale Attributed Node Embeddingcited by 548
  8. 2020Deep Graph Contrastive Representation Learningcited by 539
  9. 2020Graph Contrastive Learning with Adaptive Augmentationcited by 519
  10. 2020GCC: Graph Contrastive Coding for Graph Neural Network Pre-Trainingcited by 463
  11. 2022GraphMAE: Self-Supervised Masked Graph Autoencoderscited by 448
  12. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 397
25 more
  1. 2020Graph Structure Learning for Robust Graph Neural Networkscited by 371
  2. 2021Beyond Low-frequency Information in Graph Convolutional Networkscited by 362
  3. 2021Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methodscited by 311
  4. 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
  5. 2020Graph Representation Learning via Graphical Mutual Information Maximizationcited by 217
  6. 2021Adversarial Graph Augmentation to Improve Graph Contrastive Learningcited by 200
  7. 2019Topology Attack and Defense for Graph Neural Networks: An Optimization Perspectivecited by 196
  8. 2020GNNGuard: Defending Graph Neural Networks against Adversarial Attackscited by 195
  9. 2021Large-Scale Representation Learning on Graphs via Bootstrappingcited by 186
  10. 2022Graph Neural Networks for Graphs with Heterophily: A Surveycited by 175
  11. 2020Robust Pre-Training by Adversarial Contrastive Learningcited by 144
  12. 2021From Canonical Correlation Analysis to Self-supervised Graph Neural Networkscited by 141
  13. 2020Adversarial Self-Supervised Contrastive Learningcited by 139
  14. 2023GraphMAE2: A Decoding-Enhanced Masked Self-Supervised Graph Learnercited by 132
  15. 2020Contrastive Learning with Adversarial Examplescited by 91
  16. 2021Graph Barlow Twins: A self-supervised representation learning framework for graphscited by 85
  17. 2022Rethinking and Scaling Up Graph Contrastive Learning: An Extremely Efficient Approach with Group Discriminationcited by 69
  18. 2019Certifiable Robustness to Graph Perturbationscited by 62
  19. 2022Beyond Smoothing: Unsupervised Graph Representation Learning with Edge Heterophily Discriminatingcited by 57
  20. 2021Block Modeling-Guided Graph Convolutional Neural Networkscited by 32
  21. 2020Unsupervised Adversarially-Robust Representation Learning on Graphscited by 17
  22. 2021Graph Structural Attack by Perturbing Spectral Distancecited by 16
  23. 2025LOHA: Direct Graph Spectral Contrastive Learning Between Low-pass and High-pass Viewscited by 6
  24. 2022ARIEL: Adversarial Graph Contrastive Learningcited by 5
  25. 2023Graph Contrastive Learning under Heterophily via Graph Filterscited by 5

Built on by

Not cited yet.