Pith. sign in

REVIEW 19 cited by

Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2007.02901 v2 pith:UOYPRO74 submitted 2020-07-06 cs.LG cs.SIstat.ML

Péter Mernyei , Cătălina Cangea This is my paper · ORCID
classification cs.LGcs.SIstat.ML
keywords datasetbenchmarkdifferentexperimentsgraphnetworksneuralwiki-cs
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present Wiki-CS, a novel dataset derived from Wikipedia for benchmarking Graph Neural Networks. The dataset consists of nodes corresponding to Computer Science articles, with edges based on hyperlinks and 10 classes representing different branches of the field. We use the dataset to evaluate semi-supervised node classification and single-relation link prediction models. Our experiments show that these methods perform well on a new domain, with structural properties different from earlier benchmarks. The dataset is publicly available, along with the implementation of the data pipeline and the benchmark experiments, at https://github.com/pmernyei/wiki-cs-dataset .

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 19 Pith papers

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

  1. GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning

    cs.LG 2025-10 unverdicted novelty 7.0 of 10

    GILT turns few-shot node, edge, and graph classification into a token-reasoning problem and reaches competitive accuracy on held-out benchmarks with no per-graph tuning and no LLM.

  2. Delving into Instance-Dependent Label Noise in Graph Data: A Comprehensive Study and Benchmark

    cs.LG 2025-06 conditional novelty 7.0 of 10

    BeGIN provides 10 graph datasets with six label-noise types and a broad evaluation, showing that LLM-simulated noise hurts GNNs more than uniform or pairwise noise.

  3. Train Small, Deploy Large: Zero-Shot GNN Transfer Through Geometric Renormalization

    cs.LG 2026-07 conditional novelty 6.0 of 10

    GNN weights trained on geometric-renormalization replicas of graphs transfer zero-shot to the original graph with largely preserved accuracy.

  4. Both Topology and Text Matter: Revisiting LLM-guided Out-of-Distribution Detection on Text-attributed Graphs

    cs.LG 2026-02 conditional novelty 6.0 of 10

    LG-Plug mines pseudo-OOD exposures from clustered unlabeled nodes via iterative LLM prompting and regularizes topology-driven graph OOD detectors, cutting FPR95 by ≥7% across six TAG benchmarks.

  5. Rethinking Federated Graph Foundation Models: A Graph-Language Alignment-based Approach

    cs.LG 2026-01 conditional novelty 6.0 of 10

    FedGALA replaces vector-quantized federated graph foundation models with continuous graph-text contrastive alignment plus prompt tuning, claiming up to 14.37% gains over 22 baselines.

  6. Fixed Aggregation Features Can Rival GNNs

    cs.LG 2026-01 conditional novelty 6.0 of 10

    Fixed, untrained neighborhood statistics fed into a well-tuned MLP match or outperform classic GNNs on 12 of 14 node-classification benchmarks.

  7. Turning Tabular Foundation Models into Graph Foundation Models

    cs.LG 2025-08 conditional novelty 6.0 of 10

    G2T-FM converts graph node tasks into tabular tasks and shows that tabular foundation models can match or beat well-tuned GNNs, especially after finetuning.

  8. Quantizing Text-attributed Graphs for Semantic-Structural Integration

    cs.LG 2025-07 conditional novelty 6.0 of 10

    STAG quantizes fused graph-structure and text embeddings into discrete tokens from a frozen LLM vocabulary, enabling competitive few-shot and zero-shot node classification with or without an LLM.

  9. EVINET: Towards Open-World Graph Learning via Evidential Reasoning Network

    cs.LG 2025-06 conditional novelty 6.0 of 10

    EviNet uses Beta embeddings and subjective logic to jointly detect misclassifications and out-of-distribution nodes on graphs, outperforming baselines on five benchmarks.

  10. iN2V: Bringing Transductive Node Embeddings to Inductive Graphs

    cs.LG 2025-06 conditional novelty 6.0 of 10

    iN2V extends node2vec embeddings to unseen nodes via iterative neighbor averaging with train-time modifications, improving node classification by about 1 point on average over Feature Propagation.

  11. Graph Positional Autoencoders as Self-supervised Learners

    cs.LG 2025-05 conditional novelty 6.0 of 10

    A dual-path graph autoencoder that reconstructs node features and Laplacian-eigenvector distances reports strong self-supervised results on heterophilic and molecular benchmarks, with some overstatement in the margins...

  12. Effects of Dropout on Performance in Long-range Graph Learning Tasks

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Edge-dropping methods like DropEdge reduce sensitivity between distant nodes in GNNs, harming long-range task performance, and a sensitivity-aware variant called DropSens partially restores it.

  13. OpenRTAG: A Comprehensive Benchmark for Robust Text-Attributed Graph Learning under Data Quality Degradation

    cs.AI 2026-07 conditional novelty 5.0 of 10

    OpenRTAG is a benchmark that organizes text-attributed-graph data-quality issues into a 3x3 taxonomy (text/structure/label by sparsity/noise/imbalance) and evaluates model robustness across nine datasets and three tasks.

  14. Boundary Embedding Shaping with Adaptive Contrastive Learning for Graph Structural Disentanglement

    cs.LG 2026-06 unverdicted novelty 5.0 of 10

    Boundary-focused contrastive "gravity" loss on selected boundary nodes improves GNN node classification by about one point over an equal-architecture baseline, but the claimed proofs do not cover the implemented loss.

  15. Toward General Digraph Contrastive Learning: A Dual Spatial Perspective

    cs.LG 2025-10 conditional novelty 5.0 of 10

    S2-DiGCL combines personalized magnetic Laplacian perturbations with BFS/DFS path sampling to improve contrastive learning on directed graphs.

  16. Mitigating Degree Bias Adaptively with Hard-to-Learn Nodes in Graph Contrastive Learning

    cs.LG 2025-06 conditional novelty 5.0 of 10

    A hardness-aware reweighted contrastive loss that uses labels and similarity to upweight hard positives and negatives reduces degree bias in graph node classification.

  17. Intellectual Property in Graph-Based Machine Learning as a Service: Attacks and Defenses

    cs.CR 2025-08 conditional novelty 4.0 of 10

    A systematic review that organizes graph-ML IP protection into model-level and data-level attacks and defenses, and ships a benchmark library, PyGIP.

  18. When Noisy Labels Meet Class Imbalance on Graphs: A Graph Augmentation Method with LLM and Pseudo Label

    cs.LG 2025-07 conditional novelty 4.0 of 10

    GraphALP combines LLM-generated minority node oversampling with confidence-weighted pseudo-labeling to improve node classification on graphs that have both class imbalance and label noise.

  19. Learn Beneficial Noise as Graph Augmentation

    cs.LG 2025-05 conditional novelty 4.0 of 10

    PiNGDA learns beneficial noise on graph topology and attributes via a trainable generator, framing standard graph contrastive learning as a point estimate of positive-incentive noise.

Pith tools