mlsanityData to 7 Oct 2026

Do We Really Need Complicated Model Architectures For Temporal Networks?

arXiv 2302.11636 ↗First posted 22 Feb 2023cs.LG cs.AIcited by 141

Not on the map. Why?

Abstract (arXiv)

Recurrent neural network (RNN) and self-attention mechanism (SAM) are the de facto methods to extract spatial-temporal information for temporal graph learning. Interestingly, we found that although both RNN and SAM could lead to a good performance, in practice neither of them is always necessary. In this paper, we propose GraphMixer, a conceptually and technically simple architecture that consists of three components: (1) a link-encoder that is only based on multi-layer perceptrons (MLP) to summarize the information from temporal links, (2) a node-encoder that is only based on neighbor mean-pooling to summarize node information, and (3) an MLP-based link classifier that performs link prediction based on the outputs of the encoders. Despite its simplicity, GraphMixer attains an outstanding performance on temporal link prediction benchmarks with faster convergence and better generalization performance. These results motivate us to rethink the importance of simpler model architecture.

Key facts. Cited by 141 papers in the map (data through 7 Oct 2026). Builds on 26 of its 34 references that are in the map. First posted on arXiv 22 Feb 2023.

Builds on

26 of 34 references.

References · 26

  1. 2017Graph Attention Networkscited by 8,854
  2. 2018Graph Neural Networks: A Review of Methods and Applicationscited by 1,571
  3. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
  4. 2016Variational Graph Auto-Encoderscited by 1,373
  5. 2017Visualizing the Loss Landscape of Neural Netscited by 1,126
  6. 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,125
  7. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  8. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 624
  9. 2020Temporal Graph Networks for Deep Learning on Dynamic Graphscited by 521
  10. 2020Inductive Representation Learning on Temporal Graphscited by 454
  11. 2019EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphscited by 443
  12. 2019Predicting Dynamic Embedding Trajectory in Temporal Interaction Networkscited by 443
14 more
  1. 2019Representation Learning for Dynamic Graphs: A Surveycited by 260
  2. 2021Inductive Representation Learning in Temporal Networks via Causal Anonymous Walkscited by 188
  3. 2018dyngraph2vec: Capturing Network Dynamics using Dynamic Graph Representation Learningcited by 88
  4. 2019Variational Graph Recurrent Neural Networkscited by 86
  5. 2020APAN: Asynchronous Propagation Attention Network for Real-time Temporal Graph Embeddingcited by 78
  6. 2021Continuous-Time Sequential Recommendation with Temporal Graph Collaborative Transformercited by 69
  7. 2021TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learningcited by 65
  8. 2022TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphscited by 59
  9. 2021On Provable Benefits of Depth in Training Graph Convolutional Networkscited by 58
  10. 2022Provably expressive temporal graph networkscited by 56
  11. 2022Neighborhood-aware Scalable Temporal Network Representation Learningcited by 54
  12. 2021Time-Aware Neighbor Sampling for Temporal Graph Networkscited by 9
  13. 2021DyFormer: A Scalable Dynamic Graph Transformer with Provable Benefits on Generalization Abilitycited by 5
  14. 2022Long Short-Term Preference Modeling for Continuous-Time Sequential Recommendationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 17; 2024: 43; 2025: 41; 2026: 40.

141 papers.

Sort:

By area:

  1. 2023Towards Better Dynamic Graph Learning: New Architecture and Unified Librarycited by 154
  2. 2023Temporal Graph Benchmark for Machine Learning on Temporal Graphscited by 137
  3. 2025RelGNN: Composite Message Passing for Relational Deep Learningcited by 45
  4. 2023TempME: Towards the Explainability of Temporal Graph Neural Networks via Motif Discoverycited by 27
  5. 2024Long Range Propagation on Continuous-Time Dynamic Graphscited by 26
  6. 2024DTGB: A Comprehensive Benchmark for Dynamic Text-Attributed Graphscited by 26
  7. 2024Improving Temporal Link Prediction via Temporal Walk Matrix Projectioncited by 21
  8. 2024Efficient Neural Common Neighbor for Temporal Graph Link Predictioncited by 18
  9. 2024BACKTIME: Backdoor Attacks on Multivariate Time Series Forecastingcited by 18
  10. 2023Dynamic Graph Representation Learning with Neural Networks: A Surveycited by 15
  11. 2024Dynamic Graph Unlearning: A General and Efficient Post-Processing Method via Gradient Transformationcited by 15
  12. 2024DyG-Mamba: Continuous State Space Modeling on Dynamic Graphscited by 15
38 more
  1. 2025TGB-Seq Benchmark: Challenging Temporal GNNs with Complex Sequential Dynamicscited by 15
  2. 2024SLADE: Detecting Dynamic Anomalies in Edge Streams without Labels via Self-Supervised Learningcited by 14
  3. 2024PageRank Bandits for Link Predictioncited by 14
  4. 2024MSPipe: Efficient Temporal GNN Training via Staleness-Aware Pipelinecited by 12
  5. 2024Node-Time Conditional Prompt Learning In Dynamic Graphscited by 11
  6. 2023HOT: Higher-Order Dynamic Graph Representation Learning with Efficient Transformerscited by 10
  7. 2024Co-Neighbor Encoding Schema: A Light-cost Structure Encoding Method for Dynamic Link Predictioncited by 10
  8. 2023Spear and Shield: Adversarial Attacks and Defense Methods for Model-Based Link Prediction on Continuous-Time Dynamic Graphscited by 9
  9. 2024Self-Explainable Temporal Graph Networks based on Graph Information Bottleneckcited by 9
  10. 2024UTG: Towards a Unified View of Snapshot and Event Based Models for Temporal Graphscited by 9
  11. 2023An Empirical Evaluation of Temporal Graph Benchmarkcited by 8
  12. 2024Towards Adaptive Neighborhood for Advancing Temporal Interaction Graph Modelingcited by 8
  13. 2023EvoluNet: Advancing Dynamic Non-IID Transfer Learning on Graphscited by 7
  14. 2024Enhancing Cross-domain Link Prediction via Evolution Process Modelingcited by 7
  15. 2023Time-aware Graph Structure Learning via Sequence Prediction on Temporal Graphscited by 6
  16. 2023CAT-Walk: Inductive Hypergraph Learning via Set Walkscited by 6
  17. 2024Dynamic Graph Transformer with Correlated Spatial-Temporal Positional Encodingcited by 6
  18. 2024Retrieval Augmented Generation for Dynamic Graph Modelingcited by 6
  19. 2024Supra-Laplacian Encoding for Transformer on Dynamic Graphscited by 6
  20. 2024THeGCN: Temporal Heterophilic Graph Convolutional Networkcited by 6
  21. 2025LLM-driven Knowledge Distillation for Dynamic Text-Attributed Graphscited by 6
  22. 2025A Survey of Link Prediction in Temporal Networkscited by 6
  23. 2025Unifying Text Semantics and Graph Structures for Temporal Text-attributed Graphs with Large Language Modelscited by 6
  24. 2023Temporal graph models fail to capture global temporal dynamicscited by 5
  25. 2024Enhancing the Expressivity of Temporal Graph Networks through Source-Target Identificationcited by 5
  26. 2025Future Link Prediction Without Memory or Aggregationcited by 5
  27. 2025TIDFormer: Exploiting Temporal and Interactive Dynamics Makes A Great Dynamic Graph Transformercited by 5
  28. 2023iLoRE: Dynamic Graph Representation with Instant Long-term Modeling and Re-occurrence Preservationcited by 4
  29. 2024Learning-Based Link Anomaly Detection in Continuous-Time Dynamic Graphscited by 4
  30. 2024Temporal Graph Rewiring with Expander Graphscited by 4
  31. 2025DyGEnc: Encoding a Sequence of Textual Scene Graphs to Reason and Answer Questions in Dynamic Scenescited by 4
  32. 2025A Survey on Temporal Interaction Graph Representation Learning: Progress, Challenges, and Opportunitiescited by 4
  33. 2025What Do Temporal Graph Learning Models Learn?cited by 4
  34. 2026Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspectivecited by 4
  35. 2023GPatcher: A Simple and Adaptive MLP Model for Alleviating Graph Heterophilycited by 3
  36. 2024Input Snapshots Fusion for Scalable Discrete-Time Dynamic Graph Neural Networkscited by 3
  37. 2025UniDyG: A Unified and Effective Representation Learning Approach for Large Dynamic Graphscited by 3
  38. 2025Unlocking Multi-Modal Potentials for Link Prediction on Dynamic Text-Attributed Graphscited by 3

50 most cited of 141 All 141 (JSON)