Pith. sign in

REVIEW 4 major objections 5 minor 53 references

Revisiting Graph Contrastive Learning on Anomaly Detection: A Structural Imbalance Perspective

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Unsupervised graph anomaly detection can be made to catch low-degree tail anomalies, not just high-degree head anomalies, by forging tail nodes from head-node neighborhoods and completing tail neighborhoods from similar nodes.

desk verdict Solid empirical GAD paper with a genuine structural-imbalance angle; the core result holds, but the completion module's self-supervision loop and a few reporting overstatements need fixing. read the letter →

arxiv 2507.14677 v1 pith:GN2X7YH7 submitted 2025-07-19 cs.LG

Yiming Xu , Zhen Peng , Bin Shi , Xu Hua , Bo Dong , Song Wang , Chen Chen This is my paper · ORCID
classification cs.LG
keywords graphanomalydetectioncontrastivelearningstructuralimbalancetailnodesneighborpruningcompletionself-supervisedpower-lawnetworks
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that existing graph contrastive learning (GCL) methods for anomaly detection are structurally biased: they detect abnormal high-degree (head) nodes well but miss abnormal low-degree (tail) nodes, a failure mode that matters for real power-law networks where fraudsters and impersonators are often low-degree. To fix this, the authors propose AD-GCL, which prunes noisy edges from head nodes to create forged tail nodes, then uses knowledge learned from head nodes to supervise those forged tail nodes and guide detection of genuine tail anomalies. For tail nodes themselves, the model enlarges their small neighborhoods by mixing in ego networks of auxiliary nodes selected by both feature similarity and the model's own anomaly-score similarity. On six benchmark datasets, the paper claims AD-GCL achieves the best AUC overall and on tail and head splits, with the head-to-forged-tail alignment carrying the transfer of detection ability to real tail nodes.

What carries the argument

Two paired augmentation modules drive the method. Neighbor pruning samples $K$ neighbors for a head node $u$ from $\text{Multinomial}(K; p(\cdot|u)\cdot p_{\text{sim}}(\cdot|u))$, dropping noisy edges to forge a tail-like node. Anomaly-guided neighbor completion forms $p_{\text{mix}}(\cdot|v) = (1-\varphi)p(\cdot|v)p_{\text{nc}}(\cdot|v) + \varphi\, p(\cdot|a)p_{\text{nc}}(\cdot|a)$, where $p_{\text{nc}} = p_{\text{sim}}\cdot p_{\text{ano}}$ and the anomaly-similarity term $p_{\text{ano}}(u,v) = S_u \cdot S_v^{\top}$ uses discriminator scores from a sliding window of the past $w$ epochs; this mixes a tail node's ego network with the ego network of a similar auxiliary node. These two augmentations generate two contrastive views, and the training objective combines intra-view binary cross-entropy contrast on node-neighbor pairs with inter-view InfoNCE contrast on features and anomaly scores.

What would settle it

Build or select a benchmark where low-degree normal nodes and low-degree anomalous nodes have nearly identical features but different local connectivity, then compare anomaly-guided completion against completion using feature similarity only or random auxiliary selection; if random or feature-only completion matches AD-GCL's tail AUC, the anomaly-score guidance is not carrying the result, and if performance collapses when the first epochs' scores are frozen, the feedback loop is a liability.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that the reason GCL anomaly detectors fail on tail nodes is not insufficient model capacity but a shortage of local structure: low-degree nodes cannot generate enough diverse contrast pairs, so the discriminator overfits their sparse neighborhoods and mistakes them for normal. AD-GCL addresses this with two asymmetric augmentations. For head nodes, it samples K edges using joint neighbor-frequency and feature-saliency information so that a head node is forged into a tail-like node; aligning original head nodes with these forged tail nodes in both feature and score space transfers head-node discriminative knowledge to the tail regime. For genuine tail nodes, it samples auxiliary nodes using the product of feature similarity and an anomaly-similarity score built from a sliding window of discriminator outputs, then mixes the ego networks to enlarge the tail receptive field while preserving degree statistics. The paper reports that this scheme outperforms ten baselines on six datasets, with the largest gains on tail AUC, and that removing either augmentation degrades tail performance.

Load-bearing premise

The pipeline stands on the model's own anomaly scores from the most recent few training epochs being accurate enough to pick which neighborhoods to borrow, even though early in training those scores are noisy and the borrowed neighborhoods feed back into the same scorer.

Editorial extensions

If this is right

  • Existing GCL anomaly detectors' low tail AUC is a structural-imbalance problem, not just a capacity problem, so methods that ignore degree bias will keep missing low-degree anomalies on power-law graphs.
  • Head nodes can act as label-free teachers for tail detection: aligning original head nodes with forged tail nodes transfers discriminative knowledge without any anomaly labels.
  • Anomaly-guided neighbor completion enlarges tail receptive fields while preserving degree statistics, and the ablation study says removing it degrades tail AUC by about 1.9 percent.
  • The two augmentations and the intra/inter-view losses are complementary; removing intra-view contrast causes the largest degradation, about 30.6 percent on tail nodes, indicating it is the dominant learning signal.
  • The added modules keep the time complexity at $O(|V|d^2 + |E|d + |V|^2d)$, comparable to existing GCL anomaly detectors.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Testable extension: replace the sliding-window self-scores with a momentum encoder or exponentially smoothed scores; if early-training score noise is the limiting factor, tail performance should improve beyond the paper's windowed version.
  • On heterophilic graphs, the discriminator's neighbor-matching assumption is inverted, so the same pruning and completion recipes would likely need reversed similarity signals; the paper leaves this open.
  • The method implies a general design rule: structural imbalance in graphs can be attacked by deliberately reshaping degree distributions during augmentation rather than by reweighting the final loss.
  • A self-adaptive degree threshold could replace the fixed Pareto-based $K$ and extend the gains to graphs with very different degree distributions.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes AD-GCL, an unsupervised graph contrastive learning method for anomaly detection that targets structural imbalance. The method has two main components: neighbor pruning, which samples K salient neighbors for head nodes to create 'forged tail nodes' and aligns them with the original head nodes via inter-view contrast, and anomaly-guided neighbor completion, which enlarges the receptive field of tail nodes by mixing their ego networks with those of auxiliary nodes selected using both feature similarity and the model's own anomaly scores. The training objective combines intra-view BCE losses and inter-view InfoNCE losses. Experiments on six citation/bitcoin datasets (plus two in the appendix) report AUC for all, tail, and head nodes, and show that AD-GCL often outperforms ten baselines. The paper also provides ablations, parameter studies, complexity analysis, and a limitations appendix.

Significance. If the reported results are reliable, the paper makes a useful contribution to unsupervised graph anomaly detection by identifying a real weakness of existing GCL methods on low-degree nodes and by demonstrating a practical remedy with consistent tail-node AUC gains. The empirical study is fairly extensive: ten baselines, six main datasets, head/tail breakdowns, ablations, parameter sensitivity, and an appendix with further metrics. The authors also state that source code and datasets are released, which strengthens reproducibility. The main value is the demonstration that structural imbalance can be explicitly addressed within the GCL framework without labels. The core empirical claim is largely supported, though some overstatements and underspecified components need to be corrected.

major comments (4)
  1. [Main Results and Analysis] The claim that 'AD-GCL achieves the best anomaly detection performance on these six datasets' and 'the best AUC scores for both tail nodes and head nodes on most of the datasets' is overstated. In Table 1, on Pubmed, ANEMONE has a higher head AUC (98.20) than AD-GCL (97.88). In Appendix Table 3, on Tolokers, GAD-NR has higher overall AUPRC/AP (30.20/30.21) and higher head AUPRC/AP (31.62/31.63) than AD-GCL (27.02/27.05 and 27.14/27.19, respectively). These exceptions should be explicitly acknowledged, and the abstract/conclusion wording of 'comprehensive superiority' should be tempered to reflect the actual win/loss pattern.
  2. [Anomaly-Guided Neighbor Completion, Eq. (2)] The mixing ratio φ in Eq. (2) is not defined. The text says only that φ 'increases with the similarity to the tail node v' and is 'at most 0.5', but no formula or algorithmic specification is given. This makes the neighbor completion step irreproducible and leaves open the possibility that for highly similar auxiliary nodes φ is close to 0.5, which would substantially destroy the anchor node's own neighborhood and contradict the stated goal of enlarging the receptive field. Please provide an exact definition of φ and report the values used in the experiments, ideally with a sensitivity analysis.
  3. [Graph Contrastive Network] The training schedule is ambiguous. The text says that during the initial stage view1 is the original graph and view2 is the neighbor-pruned graph, and that 'in the later stage of training, we use the neighbor completion strategy ... generating two augmented graphs (referred to as view1 and view2)'. It is unclear (a) at what epoch or by what criterion the switch occurs, (b) whether view1 in the later stage remains the original graph or is also the result of completion, and (c) which view(s) are used when computing the anomaly scores S_{v} that feed into p_ano. This underspecification affects both reproducibility and the interpretation of the inter-view contrastive loss.
  4. [Anomaly-Guided Neighbor Completion and Parameter Study] The evidence that anomaly-guided selection, rather than the mixup augmentation itself, drives the tail-node improvement is incomplete. The distribution p_ano is computed from the discriminator's own scores over a sliding window; early in training these scores are noisy, and because the completed neighborhoods are fed back into the same discriminator, selection errors could be amplified. The w=0 ablation in Figure 4d is only shown for Cora and does not isolate feedback effects. Please add control experiments, for example sampling auxiliary nodes by feature similarity alone (without the p_ano term) and reporting the w=0 ablation on all datasets, or providing an analysis of how p_ano evolves during training.
minor comments (5)
  1. [Anomaly Score Calculation, Eq. (10)] The symbol S is used both for the sliding-window anomaly-score matrix in the completion section and for the final anomaly score in Eq. (10); please use distinct notations to avoid confusion.
  2. [Neighbor Pruning Strategy, Eq. (1)] The product p(·|u) · psim(·|u) is described as sampling from a multinomial distribution; please state explicitly that the product is normalized to form a valid probability distribution, or define Multinomial to accept unnormalized weights.
  3. [Ablation Study] The numerical degradation percentages (2.46%, 1.87%, 30.62%, 1.89%) reported in the text are not tied to a specific table or figure; please include the underlying values in Figure 3 or a supplementary table for reproducibility.
  4. [Appendix, Table 3] The main text consistently refers to 'six datasets', while the appendix reports results on eight datasets; please harmonize the dataset counts and clarify that Table 3 covers a subset of datasets with AUPRC/AP.
  5. [Limitations] The limitations section is a welcome addition; consider also mentioning that the anomaly-guided completion depends on the quality of the discriminator's early training scores, which is currently not analyzed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the central claims are benchmark measurements, and the self-referential training signal does not reduce to the test-time prediction.

full rationale

The paper is an empirical method paper. Its central claim is that AD-GCL achieves the best AUC on six benchmark datasets, which is established by direct measurement against ten baselines, not derived from the method's definitions. The only self-referential element is the anomaly-guided neighbor completion: the model uses the discriminator's own training-time scores S to define p_ano(u,v) = S_u · S_v^T, and then trains the discriminator on the completed neighborhoods. This is a bootstrap-style augmentation, but it is not circular in the sense required here. The final anomaly score in Eq. (10) is computed at inference on the original graph from positive and negative discriminator scores, so it is not by construction equal to the training-time selection scores. The w=0 ablation in Figure 4d shows that removing the anomaly-guidance term causes a significant performance drop, and Table 4 shows that naive edge completion degrades performance, providing non-vacuous checks that the completion component contributes beyond simply adding edges. No fitted parameter is renamed as a prediction: the tunable hyperparameters (w, alpha, R, d) are studied via parameter experiments, and none of the paper's claims is an algebraic identity with them. The self-citations to the authors' prior works (e.g., Xu et al. 2023b, 2024) are used only for standard readout choices and related-work context, and are not load-bearing for the anomaly detection results. The paper also acknowledges its limitations regarding heterophilic graphs and the lack of theoretical bounds, which further indicates that the claims are empirical rather than disguised derivations. Consequently, no circular step can be exhibited from the paper's own equations or self-citations.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The paper leans on standard contrastive assumptions (homophily, informative neighborhoods) and on several hand-chosen parameters that directly shape the head/tail evaluation, so the central claim is an empirical method result rather than a derived theorem.

free parameters (6)
  • Degree threshold K = 6 (Cora, Citeseer, Pubmed, Bitcoinotc, BITotc, BITalpha), 9 (Reddit), 90 (Tolokers)
    Defines the head/tail split and the number of sampled neighbors in pruning; set by Pareto principle per Appendix, affects all reported head/tail AUCs.
  • Sliding window w = 5
    Selected because Figure 4d shows best test AUC on Cora at w=5; controls how many past epochs of self-scores feed neighbor completion.
  • Trade-off parameter alpha = 0.2
    Selected from Figure 4c test-curve on Cora; balances intra- and inter-view losses.
  • Mixing ratio phi = unspecified function of similarity, capped at 0.5
    Controls the mixup in Eq. (2); no formula given, only described qualitatively, so the completion step is underspecified and hard to reproduce exactly.
  • Sampling rounds R = 256
    Selected from Figure 4a; number of rounds in anomaly scoring Eq. (10).
  • Hidden dimension d = 64
    Selected from Figure 4b; standard hyperparameter.
assumptions (4)
  • domain assumption Normal nodes match their neighbors; anomalous nodes deviate from local patterns (homophily)
    Used throughout the graph contrastive network to define positive and negative pairs; acknowledged as a limitation for heterophilic graphs in the Appendix.
  • domain assumption Real-world graphs follow power-law degree distributions
    Motivates the head/tail dichotomy; supported by degree histograms in Figure A5.
  • domain assumption Random walk with restart yields representative local subgraphs for contrast pairs
    Imported from CoLA and earlier work; used in Eq. (4) to sample neighbor sets.
  • domain assumption Sliding-window discriminator scores are informative for selecting auxiliary nodes
    The completion sampler in Eq. (2) uses p_ano from the model's own scores; only the w=0 ablation supports this, with no formal guarantee.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Revisiting Graph Contrastive Learning on Anomaly Detection: A Structural Imbalance Perspective." pith.science (2026). https://pith.science/paper/GN2X7YH7

@misc{pith2026250714677,
  author       = {Pith},
  title        = {Pith review of: Revisiting Graph Contrastive Learning on Anomaly Detection: A Structural Imbalance Perspective},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GN2X7YH7}},
  note         = {Machine review of arXiv:2507.14677}
}
read the original abstract

The superiority of graph contrastive learning (GCL) has prompted its application to anomaly detection tasks for more powerful risk warning systems. Unfortunately, existing GCL-based models tend to excessively prioritize overall detection performance while neglecting robustness to structural imbalance, which can be problematic for many real-world networks following power-law degree distributions. Particularly, GCL-based methods may fail to capture tail anomalies (abnormal nodes with low degrees). This raises concerns about the security and robustness of current anomaly detection algorithms and therefore hinders their applicability in a variety of realistic high-risk scenarios. To the best of our knowledge, research on the robustness of graph anomaly detection to structural imbalance has received little scrutiny. To address the above issues, this paper presents a novel GCL-based framework named AD-GCL. It devises the neighbor pruning strategy to filter noisy edges for head nodes and facilitate the detection of genuine tail nodes by aligning from head nodes to forged tail nodes. Moreover, AD-GCL actively explores potential neighbors to enlarge the receptive field of tail nodes through anomaly-guided neighbor completion. We further introduce intra- and inter-view consistency loss of the original and augmentation graph for enhanced representation. The performance evaluation of the whole, head, and tail nodes on multiple datasets validates the comprehensive superiority of the proposed AD-GCL in detecting both head anomalies and tail anomalies.

Figures

Figures reproduced from arXiv: 2507.14677 by the authors.

Figure 1
Figure 1. An illustrative showcasing performance disparity between head and tail anomalies. Blue scatters represent the AUC at [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. (a) The architecture of the AD-GCL; (b) Neighbor pruning to filter noise edges of head nodes; (c) Anomaly-guided [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Ablation study on different variants. 1 2 4 16 32 64 128 256 512 0.75 0.80 0.85 0.90 0.95 AUC AUC TN HN (a) Sampling rounds R 1 2 4 8 16 32 64 128 256 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 AUC AUC TN HN (b) Dimension d 0.005 0.01 0.05 0.1 0.2 0.3 0.4 0.5 1 0.80 0.85 0.90 0.95 AUC AUC TN HN (c) Trade-off parameter α 0 5 10 15 20 25 30 35 40 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 AUC AUC TN HN (d) Sliding window w … view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: The parameter study of AD-GCL with varying (a) sampling rounds [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 47 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Bandyopadhyay, S.; Lokesh, N.; and Murty, M. N. 2019. Outlier aware network embedding for attributed networks. In Proceedings of the AAAI conference on artificial intelligence, volume 33, 12--19

    work page 2019

  4. [4]

    V.; and Murty, M

    Bandyopadhyay, S.; N, L.; Vivek, S. V.; and Murty, M. N. 2020. Outlier resistant unsupervised deep architectures for attributed network embedding. In Proceedings of the 13th international conference on web search and data mining, 25--33

    work page 2020

  5. [5]

    Chen, J.; Zhu, G.; Yuan, C.; and Huang, Y. 2024. Boosting Graph Anomaly Detection with Adaptive Message Passing. In The Twelfth International Conference on Learning Representations

    work page 2024

  6. [6]

    Chen, Z.; Liu, B.; Wang, M.; Dai, P.; Lv, J.; and Bo, L. 2020. Generative adversarial attributed network anomaly detection. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 1989--1992

    work page 2020

  7. [7]

    Ding, K.; Li, J.; Agarwal, N.; and Liu, H. 2021. Inductive anomaly detection on attributed networks. In Proceedings of the twenty-ninth international conference on international joint conferences on artificial intelligence, 1288--1294

    work page 2021

  8. [8]

    Ding, K.; Li, J.; Bhanushali, R.; and Liu, H. 2019. Deep anomaly detection on attributed networks. In SDM, 594--602. SIAM

    work page 2019

Show all 53 references
  1. [9]

    Dou, Y.; Liu, Z.; Sun, L.; Deng, Y.; Peng, H.; and Yu, P. S. 2020. Enhancing graph neural network-based fraud detectors against camouflaged fraudsters. In Proceedings of the 29th ACM international conference on information & knowledge management, 315--324

    2020

  2. [10]

    Duan, J.; Wang, S.; Zhang, P.; Zhu, E.; Hu, J.; Jin, H.; Liu, Y.; and Dong, Z. 2023. Graph anomaly detection via multi-scale contrastive learning networks with augmented view. In AAAI, volume 37, 7459--7467

    2023

  3. [11]

    Goga, O.; Venkatadri, G.; and Gummadi, K. P. 2015. The doppelg \"a nger bot attack: Exploring identity impersonation in online social networks. In Proceedings of the 2015 internet measurement conference, 141--153

    2015

  4. [12]

    He, K.; Fan, H.; Wu, Y.; Xie, S.; and Girshick, R. 2020. Momentum contrast for unsupervised visual representation learning. In CVPR, 9729--9738

    2020

  5. [13]

    A.; Beutel, A.; Shah, N.; and Faloutsos, C

    Hooi, B.; Shin, K.; Song, H. A.; Beutel, A.; Shah, N.; and Faloutsos, C. 2017. Graph-based fraud detection in the face of camouflage. ACM Transactions on Knowledge Discovery from Data (TKDD), 11(4): 1--26

    2017

  6. [14]

    Hu, J.; Xiao, B.; Jin, H.; Duan, J.; Wang, S.; Lv, Z.; Wang, S.; Liu, X.; and Zhu, E. 2023. SAMCL: Subgraph-Aligned Multiview Contrastive Learning for Graph Anomaly Detection. IEEE Transactions on Neural Networks and Learning Systems

    2023

  7. [15]

    Jin, M.; Liu, Y.; Zheng, Y.; Chi, L.; Li, Y.-F.; and Pan, S. 2021. Anemone: Graph anomaly detection with multi-scale contrastive learning. In CIKM, 3122--3126

    2021

  8. [16]

    P.; and Ba, J

    Kingma, D. P.; and Ba, J. 2015. Adam: A method for stochastic optimization

    2015

  9. [17]

    N.; and Welling, M

    Kipf, T. N.; and Welling, M. 2017. Semi-supervised classification with graph convolutional networks

    2017

  10. [18]

    Kumar, S.; Hooi, B.; Makhija, D.; Kumar, M.; Faloutsos, C.; and Subrahmanian, V. 2018. Rev2: Fraudulent user prediction in rating platforms. In Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, 333--341

    2018

  11. [19]

    Kumar, S.; Spezzano, F.; Subrahmanian, V.; and Faloutsos, C. 2016. Edge weight prediction in weighted signed networks. In ICDM, 221--230. IEEE

    2016

  12. [20]

    Li, J.; Dani, H.; Hu, X.; and Liu, H. 2017. Radar: Residual analysis for anomaly detection in attributed networks. In IJCAI, volume 17, 2152--2158

    2017

  13. [21]

    Liao, J.; Li, J.; Chen, L.; Wu, B.; Bian, Y.; and Zheng, Z. 2023. SAILOR: Structural Augmentation Based Tail Node Representation Learning. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, 1389--1399

    2023

  14. [22]

    Liu, K.; Dou, Y.; Zhao, Y.; Ding, X.; Hu, X.; Zhang, R.; Ding, K.; Chen, C.; Peng, H.; Shu, K.; et al. 2022. Bond: Benchmarking unsupervised outlier node detection on static attributed graphs. Advances in Neural Information Processing Systems, 35: 27021--27035

    2022

  15. [23]

    Liu, Y.; Ao, X.; Qin, Z.; Chi, J.; Feng, J.; Yang, H.; and He, Q. 2021 a . Pick and choose: a GNN-based imbalanced learning approach for fraud detection. In Proceedings of the web conference 2021, 3168--3177

    2021

  16. [24]

    Liu, Y.; Li, Z.; Pan, S.; Gong, C.; Zhou, C.; and Karypis, G. 2021 b . Anomaly detection on attributed networks via contrastive self-supervised learning. TNNLS, 33(6): 2378--2392

    2021

  17. [25]

    Liu, Z.; Li, Y.; Chen, N.; Wang, Q.; Hooi, B.; and He, B. 2023. A survey of imbalanced learning on graphs: Problems, techniques, and future directions. arXiv preprint arXiv:2308.13821

    2023 arXiv

  18. [26]

    Liu, Z.; Nguyen, T.-K.; and Fang, Y. 2021. Tail-gnn: Tail-node graph neural networks. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, 1109--1119

    2021

  19. [27]

    Liu, Z.; Zhang, W.; Fang, Y.; Zhang, X.; and Hoi, S. C. 2020. Towards locality-aware meta-learning of tail node embeddings on networks. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 975--984

    2020

  20. [28]

    Z.; Xiong, H.; and Akoglu, L

    Ma, X.; Wu, J.; Xue, S.; Yang, J.; Zhou, C.; Sheng, Q. Z.; Xiong, H.; and Akoglu, L. 2021. A comprehensive survey on graph anomaly detection with deep learning. TKDE

    2021

  21. [29]

    Pacini, C.; Hopwood, W.; Young, G.; and Crain, J. 2018. The role of shell entities in fraud and other financial crimes. Managerial Auditing Journal, 34(3): 247--267

    2018

  22. [30]

    Park, J.; Song, J.; and Yang, E. 2022. Graphens: Neighbor-aware ego network synthesis for class-imbalanced node classification. In The Tenth International Conference on Learning Representations, ICLR 2022. International Conference on Learning Representations (ICLR)

    2022

  23. [31]

    D.; Noghre, G

    Pazho, A. D.; Noghre, G. A.; Purkayastha, A. A.; Vempati, J.; Martin, O.; and Tabkhi, H. 2023. A Survey of Graph-based Deep Learning for Anomaly Detection in Distributed Systems. IEEE Transactions on Knowledge and Data Engineering

    2023

  24. [32]

    W.; Francis, M

    Pennebaker, J. W.; Francis, M. E.; and Booth, R. J. 2001. Linguistic inquiry and word count: LIWC 2001. Mahway: Lawrence Erlbaum Associates, 71(2001): 2001

    2001

  25. [33]

    Platonov, O.; Kuznedelev, D.; Diskin, M.; Babenko, A.; and Prokhorenkova, L. 2023. A critical look at the evaluation of GNNs under heterophily: Are we really making progress? ICLR

    2023

  26. [34]

    Roy, A.; Shu, J.; Li, J.; Yang, C.; Elshocht, O.; Smeets, J.; and Li, P. 2023. GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction. Proceedings of the 17th ACM International Conference on Web Search and Data Mining (WSDM)

    2023

  27. [35]

    Sanders, R. 1987. The Pareto principle: its use and abuse. Journal of Services Marketing, 1(2): 37--40

    1987

  28. [36]

    Skillicorn, D. B. 2007. Detecting anomalies in graphs. In 2007 IEEE Intelligence and Security Informatics, 209--216. IEEE

    2007

  29. [37]

    Song, X.; Wu, M.; Jermaine, C.; and Ranka, S. 2007. Conditional anomaly detection. IEEE Transactions on knowledge and Data Engineering, 19(5): 631--645

    2007

  30. [38]

    Tang, J.; Hua, F.; Gao, Z.; Zhao, P.; and Li, J. 2024. Gadbench: Revisiting and benchmarking supervised graph anomaly detection. Advances in Neural Information Processing Systems, 36

    2024

  31. [39]

    Tang, X.; Yao, H.; Sun, Y.; Wang, Y.; Tang, J.; Aggarwal, C.; Mitra, P.; and Wang, S. 2020. Investigating and mitigating degree-related biases in graph convoltuional networks. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 1435--1444

    2020

  32. [40]

    Tong, H.; Faloutsos, C.; and Pan, J.-Y. 2006. Fast random walk with restart and its applications. In Sixth international conference on data mining (ICDM'06), 613--622. IEEE

    2006

  33. [41]

    Van den Oord, A.; Li, Y.; and Vinyals, O. 2018. Representation learning with contrastive predictive coding. arXiv e-prints, arXiv--1807

    2018

  34. [42]

    L.; Li \`o , P.; Bengio, Y.; and Hjelm, R

    Veli c kovi \'c , P.; Fedus, W.; Hamilton, W. L.; Li \`o , P.; Bengio, Y.; and Hjelm, R. D. 2019. Deep graph infomax. ICLR

    2019

  35. [43]

    Wang, R.; Wang, X.; Shi, C.; and Song, L. 2022. Uncovering the Structural Fairness in Graph Contrastive Learning. Advances in Neural Information Processing Systems, 35: 32465--32473

    2022

  36. [44]

    Wu, J.; He, J.; and Xu, J. 2019. Net: Degree-specific graph neural networks for node and graph classification. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 406--415

    2019

  37. [45]

    Wu, T.; Ren, H.; Li, P.; and Leskovec, J. 2020. Graph information bottleneck. Advances in Neural Information Processing Systems, 33: 20437--20448

    2020

  38. [46]

    Xu, H.; Xiang, L.; Huang, F.; Weng, Y.; Xu, R.; Wang, X.; and Zhou, C. 2023 a . Grace: Graph Self-Distillation and Completion to Mitigate Degree-Related Biases. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2813--2824

    2023

  39. [47]

    Xu, X.; Yuruk, N.; Feng, Z.; and Schweiger, T. A. 2007. Scan: a structural clustering algorithm for networks. In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, 824--833

    2007

  40. [48]

    Xu, Y.; Peng, Z.; Shi, B.; Hua, X.; and Dong, B. 2024. Learning dynamic graph representations through timespan view contrasts. Neural Networks, 176: 106384

    2024

  41. [49]

    Xu, Y.; Shi, B.; Ma, T.; Dong, B.; Zhou, H.; and Zheng, Q. 2023 b . CLDG: Contrastive Learning on Dynamic Graphs. In 2023 IEEE 39th International Conference on Data Engineering (ICDE), 696--707. IEEE

    2023

  42. [50]

    Ye, G.; Yin, H.; Chen, T.; Chen, H.; Cui, L.; and Zhang, X. 2021. FENet: a frequency extraction network for obstructive sleep apnea detection. IEEE Journal of Biomedical and Health Informatics, 25(8): 2848--2856

    2021

  43. [51]

    Zeng, L.; Li, L.; Gao, Z.; Zhao, P.; and Li, J. 2023. Imgcl: Revisiting graph contrastive learning on imbalanced node classification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 11138--11146

    2023

  44. [52]

    Zhang, J.; and et al. 2022. Reconstruction enhanced multi-view contrastive learning for anomaly detection on attributed networks. IJCAI

    2022

  45. [53]

    Zheng, Q.; Xu, Y.; Liu, H.; Shi, B.; Wang, J.; and Dong, B. 2023. A Survey of Tax Risk Detection Using Data Mining Techniques. Engineering

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.