Pith. sign in

REVIEW 24 cited by

SIGN: Scalable Inception Graph Neural Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2004.11198 v3 pith:MKOUJY4I submitted 2020-04-23 cs.LG stat.ML

classification cs.LGstat.ML
keywords graphnetworkstrainingarchitecturedifferentefficientinferencelearning
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Graph representation learning has recently been applied to a broad spectrum of problems ranging from computer graphics and chemistry to high energy physics and social media. The popularity of graph neural networks has sparked interest, both in academia and in industry, in developing methods that scale to very large graphs such as Facebook or Twitter social networks. In most of these approaches, the computational cost is alleviated by a sampling strategy retaining a subset of node neighbors or subgraphs at training time. In this paper we propose a new, efficient and scalable graph deep learning architecture which sidesteps the need for graph sampling by using graph convolutional filters of different size that are amenable to efficient precomputation, allowing extremely fast training and inference. Our architecture allows using different local graph operators (e.g. motif-induced adjacency matrices or Personalized Page Rank diffusion matrix) to best suit the task at hand. We conduct extensive experimental evaluation on various open benchmarks and show that our approach is competitive with other state-of-the-art architectures, while requiring a fraction of the training and inference time. Moreover, we obtain state-of-the-art results on ogbn-papers100M, the largest public graph dataset, with over 110 million nodes and 1.5 billion edges.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 24 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. When does distribution shift break graph neural networks calibration?

    cs.LG 2026-07 conditional novelty 7.0 of 10

    GNN calibration under distribution shift is governed by a single closed-form slope κ(hs, ht, ρ) that sets the optimal global temperature T⋆=1/κ and explains when node-wise recalibration cannot help.

  2. OpenGU: A Comprehensive Benchmark for Graph Unlearning

    cs.LG 2025-01 conditional novelty 7.0 of 10

    OpenGU is claimed as the first comprehensive benchmark for graph unlearning, integrating 16 methods, 37 datasets, and a 3x3 grid of tasks and unlearning requests, with empirical findings on effectiveness, efficiency, ...

  3. HOPPER: Learnable Hop Extraction for Linearized Graph Sequence Models

    cs.LG 2026-08 conditional novelty 6.0 of 10

    HOPPER learns graph- and hop-adaptive sequence extraction for linearized graph sequence models, recovering fixed LGSM extractors as special cases and achieving the best reported scores on eccentricity and shortest-pat...

  4. MiGHT-EHR: A Multi-task Graph Transformer for Heterogeneous Temporal Electronic Health Records

    cs.LG 2026-08 reject novelty 6.0 of 10

    A heterogeneous graph transformer with temporal attention and balanced multi-task training reports state-of-the-art average performance on four EHR prediction tasks on MIMIC-III and MIMIC-IV, but the evaluation may le...

  5. No Need to Train Your RDB Foundation Model

    cs.AI 2026-02 conditional novelty 6.0 of 10

    Column-wise, parameter-free JUICE encodings let single-table ICL models solve multi-table RDB prediction tasks with no training or fine-tuning.

  6. FedSA-GCL: A Semi-Asynchronous Federated Graph Learning Framework with Personalized Aggregation and Cluster-Aware Broadcasting

    cs.LG 2025-07 conditional novelty 6.0 of 10

    A semi-asynchronous federated graph learning framework with soft-label clustering, staleness-weighted aggregation, and cluster broadcasting reports higher accuracy and faster convergence than ten baselines.

  7. HOPSE: Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations

    cs.LG 2025-05 conditional novelty 6.0 of 10

    HOPSE encodes higher-order topological data by applying graph positional and structural encoders to Hasse graph decompositions, matching or exceeding message-passing models on benchmarks with up to 7x faster training.

  8. Graph Learning at Scale: Characterizing and Optimizing Pre-Propagation GNNs

    cs.LG 2025-04 conditional novelty 6.0 of 10

    Pre-propagation GNNs, whose training is bottlenecked by data loading, can be sped up roughly 15x with custom loaders, GPU double buffering, chunk reshuffling, and direct storage access, beating sampling-based GNNs by ...

  9. Even Sparser Graph Transformers

    cs.LG 2024-11 conditional novelty 6.0 of 10

    Spexphormer trains a narrow graph transformer to identify important attention edges, then trains a wider model on the sparsified graph, achieving competitive accuracy with far less memory.

  10. Personalized One-shot Federated Graph Learning for Heterogeneous Clients

    cs.LG 2024-11 conditional novelty 6.0 of 10

    O-pFGL achieves one-shot personalized federated graph learning by aggregating class-wise feature statistics into a surrogate graph and combining global distillation with local fine-tuning, outperforming baselines on 1...

  11. IMPaCT GNN: Imposing invariance with Message Passing in Chronological split Temporal Graphs

    cs.LG 2024-11 reject novelty 6.0 of 10

    IMPaCT reweights graph message passing to align first and second moments across chronological train/test splits, reporting accuracy gains on ogbn-mag and a theoretical W1 bound.

  12. Efficiently Learning Branching Networks for Multitask Algorithmic Reasoning

    cs.LG 2025-11 conditional novelty 5.0 of 10

    AutoBRANE learns tree-structured branching networks for multitask algorithmic reasoning via gradient-based task affinities and convex relaxation.

  13. HIAL: A New Paradigm for Hypergraph Active Learning via Influence Maximization

    cs.LG 2025-07 conditional novelty 5.0 of 10

    HIAL selects labels for hypergraph neural networks by maximizing a dual feature-and-topology influence function, and reports improved accuracy on seven datasets.

  14. Uncertainty-Aware Graph Neural Networks: A Multi-Hop Evidence Fusion Approach

    cs.LG 2025-06 conditional novelty 5.0 of 10

    EFGNN fuses per-depth evidential opinions from a multi-hop GNN into one final Dirichlet-based prediction whose uncertainty is lower than that of any single propagation depth.

  15. Transaction Categorization with Relational Deep Learning in QuickBooks

    cs.CE 2025-06 conditional novelty 5.0 of 10

    Rel-Cat predicts transaction categories by converting the QuickBooks relational database into a heterogeneous graph and treating categorization as link prediction, beating production baselines on a private dataset.

  16. A Generative Graph Contrastive Learning Model with Global Signal

    cs.LG 2025-04 conditional novelty 5.0 of 10

    A graph contrastive learning framework that uses a low-rank SVD-smoothed adjacency matrix as the augmented view and adaptively reweights hard and easy contrastive pairs to improve node classification.

  17. ScaDyG:A New Paradigm for Large-scale Dynamic Graph Learning

    cs.LG 2025-01 reject novelty 5.0 of 10

    ScaDyG proposes a decoupled, step-wise exponential time encoding with hypernetwork aggregation for scalable dynamic graph learning, but the key proof of equivalence with composite exponential message passing is invali...

  18. Toward Scalable Graph Unlearning: A Node Influence Maximization based Approach

    cs.LG 2025-01 conditional novelty 5.0 of 10

    The paper's NIM+SGU pipeline improves forgetting and preserves accuracy in graph unlearning by selecting high-influence nodes via propagation-based influence scores and fine-tuning on entity-specific losses.

  19. Toward Effective Digraph Representation Learning: A Magnetic Adaptive Propagation based Approach

    cs.LG 2025-01 conditional novelty 5.0 of 10

    Replacing the hand-tuned global q of magnetic digraph GNNs with an automatically computed per-edge q improves accuracy on 12 directed-graph benchmarks, with a trainable variant reaching reported SOTA.

  20. Large-Scale Spectral Graph Neural Networks via Laplacian Sparsification: Technical Report

    cs.LG 2025-01 reject novelty 5.0 of 10

    SGNN-LS approximates dense multi-hop spectral filters with sparse random-walk graphs, enabling end-to-end training on 111M-node graphs, but the theory for negative filter coefficients has a gap.

  21. Towards Scalable and Deep Graph Neural Networks via Noise Masking

    cs.LG 2024-12 reject novelty 5.0 of 10

    RMask uses random walks and a distance-based mask to extract exact-hop graph information, improving the accuracy and speed of model-simplification GNNs.

  22. SCFCRC: Simultaneously Counteract Feature Camouflage and Relation Camouflage for Fraud Detection

    cs.LG 2025-01 conditional novelty 4.0 of 10

    SCFCRC combines a feature filter trained with propagated pseudo-labels and a mixture-of-experts refiner with masking regularization to improve fraud detection on YelpChi and Amazon.

  23. Towards Data-centric Machine Learning on Directed Graphs: a Survey

    cs.LG 2024-11 unverdicted novelty 3.0 of 10

    A survey taxonomizing directed graph neural networks into message-passing, eigenpolynomial, and sequence-based frameworks and re-reading them from a data-centric perspective.

  24. Topological Signal Processing and Learning: Recent Advances and Future Challenges

    eess.SP 2024-12 accept novelty 2.0 of 10

    A survey of topological signal processing and learning that uses Hodge theory to unify recent advances in processing signals on simplicial and cell complexes.

Pith tools