Pith. sign in

REVIEW 4 major objections 4 minor 166 references

This survey claims that GNN-based link prediction is best understood through a two-dimensional taxonomy: four backbone architectures (GCN, GAE, GAT, GFormer) and two application areas (knowledge graphs and recommender systems), making the s

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 15:14 UTC pith:CI7AUTLP

load-bearing objection A useful survey map of GNN-based link prediction, but its four-way taxonomy doesn't cleanly partition the field because GAE is a wrapper around GCN and the other categories sit at different levels of abstraction. the 4 major comments →

arxiv 2607.16198 v1 pith:CI7AUTLP submitted 2026-04-29 cs.AI cs.LGcs.SI

A Survey on GNN-based Link Prediction: Techniques, Applications, and Challenges

Chengcheng Sun , Yajie Song , Cheng Zhai , Jiayun Tian , Jia Yang , Xiaobin Rui , Jian Zhang , Zhixiao Wang
show 1 more author
Philip S. Yu
This is my paper
classification cs.AI cs.LGcs.SI
keywords link predictiongraph neural networksGCNgraph autoencodersgraph attention networksgraph transformersknowledge graph completionsession-based recommendation
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This survey tries to establish that the fast-growing body of GNN-based link prediction research can be coherently organized by a two-dimensional taxonomy: technique (which GNN backbone a model uses) and application (knowledge-graph completion or recommendation). It argues that earlier reviews either treated link prediction as a generic downstream task or focused on specific network types, leaving no systematic map of the GNN architectures built specifically for edge inference. The payoff of the taxonomy is prescriptive: by attaching strengths and limitations to each of the four backbone families, the survey aims to guide practitioners in choosing an architecture for a given graph type and task. A sympathetic reader would take the main contribution to be the organizing frame itself, not a new technical result.

Core claim

On its own terms, the paper's central claim is that previous link prediction surveys are fragmented, and that a unified 'GNN perspective' fills the gap: it categorizes recent methods into GCN-based (node-wise, pair-wise, and position-encoding), GAE-based (unsupervised reconstruction), GAT-based (local and global attention), and GFormer-based (global self-attention) families, then maps those onto knowledge-graph completion and reasoning and onto session-based and personalized recommendation. The survey further claims that this architecture-centric framing reveals a methodological evolution from local neighborhood aggregation to global dependency modeling, and that comparing the four families

What carries the argument

The central organizing device is the two-dimensional taxonomy (techniques × applications). On the technique axis, the four backbone families are defined by their message-aggregation mechanism: GCN aggregates neighbor features with a normalized adjacency matrix; GAE learns node embeddings by reconstructing the adjacency matrix; GAT weights neighbors by learned attention coefficients; GFormer applies transformer self-attention to graph structure. On the application axis, knowledge-graph link prediction (completion, inductive reasoning, complex query answering) and recommender-system link prediction (session-based recommendation as a special case) anchor the real-world deployment discussion. Th

Load-bearing premise

The load-bearing premise is that the four-way backbone taxonomy (GCN, GAE, GAT, GFormer) is a valid and complete way to partition GNN-based link prediction; if important methods do not fit cleanly into these categories, or if the categories overlap, the survey's central organizing contribution weakens.

What would settle it

A concrete test: take a random sample of 100 recent GNN-based link prediction papers and check whether each can be assigned unambiguously to exactly one of the four categories. If a large fraction (say, more than 20 percent) falls outside the taxonomy or straddles multiple categories, the claim of a complete two-dimensional framework fails. Additionally, locating an earlier review that already systematically surveys link prediction from a dedicated GNN perspective would falsify the 'one of the first' assertion.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the taxonomy is right, the relevant design choice for a link-prediction practitioner is the backbone family: GCN and GAE for homogeneous and large-scale graphs, GAT for heterogeneous, dynamic, or noisy graphs, GFormer when long-range dependencies matter.
  • The survey's comparison implies that no single architecture dominates: GCN suffers from over-smoothing, GAE from limited expressiveness and transductive bias, and GAT and GFormer from computational cost, so model selection should be driven by graph type and scale.
  • The open challenges named in the paper—complex graph structures, structural feature expressiveness, self-supervised learning, and scalability—become the agenda for the next generation of GNN link predictors.
  • Treating session-based recommendation as a link prediction problem means advances in GNN link prediction can transfer directly to recommender systems, and vice versa.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A reader might infer that the taxonomy's 'GFormer-based' bucket is the most fluid: as graph transformers absorb attention and position-encoding ideas from the other families, the boundaries between the four categories may blur in future work.
  • The survey's emphasis on distinguishing isomorphic links points to a testable benchmark: comparing GCN-based against GFormer-based methods on graphs with many isomorphic node pairs would directly probe the expressiveness gap the paper describes.
  • One consequence the authors leave implicit is that the same encoder-decoder pipeline (GNN encoder plus MLP decoder) could make GNN-based link prediction a drop-in component for graph foundation models, since it already separates representation from prediction.
  • The paper's challenge list suggests that scalability, not accuracy alone, is the binding constraint; a practical extension would be to measure the four backbone families on a common large-scale dynamic graph benchmark.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper is a survey of GNN-based link prediction. Its stated contribution is a two-dimensional taxonomy: a technique axis with four categories (GCN-based, GAE-based, GAT-based, GFormer-based) and an application axis covering knowledge graphs and recommender systems. The survey describes preliminaries and graph types, summarizes representative methods in Table 1, discusses each technique family and both application domains, and closes with challenges (complex graphs, structural features, self-supervised learning, scalability). The central claim is that this is among the first reviews to organize link prediction from a dedicated GNN-architecture perspective and to offer a prescriptive, comparative framework.

Significance. If the proposed taxonomy were sound, the survey would fill a genuine gap: prior reviews treat GNNs as generic encoders or focus on classic link prediction, while this paper attempts to organize methods by architectural backbone. The manuscript has a broad reference list, a public GitHub repository, and a useful discussion of applications and open problems. These are real strengths. However, the central taxonomy is internally inconsistent: the four technique categories are not mutually exclusive, not all at the same level of abstraction, and are applied in ways that misclassify or mischaracterize particular methods. Because the taxonomy is the paper's main claimed novelty, the contribution needs substantial repair rather than minor polishing.

major comments (4)
  1. [Section 3, Table 1, Eq. (4)] The four-way technique taxonomy is not a valid partition. Section 2.2 defines GAE as an encoder–decoder whose encoder is exactly GCN (Eq. 4: Z = GCN(X,A)), so any GAE-based model is also GCN-based by construction. The categories therefore overlap, and Table 1 assigns methods to one bucket only: Labeling Trick and NCNC are listed as GAE-based although their core mechanisms are a labeling scheme and common-neighbor pooling, while xGCN is listed under GCN despite being unsupervised. GAT is a layer architecture, GFormer is a broad hybrid family, and GAE is a training objective/decoder; these are not comparable axes. The text says the categories are 'based on their backbone networks' and presents the taxonomy as fine-grained and prescriptive. The authors should either redefine the taxonomy as multi-label/non-exclusive or justify the chosen grouping as a pragmatic clustering; otherwise the cen
  2. [Section 4.1, Table 1] The description of Ran et al. (2024) is inconsistent with the cited paper. Table 1 labels the entry DPLP under GCN-based pair-wise methods, and the text states that 'Ran et al. propose an innovative path subgraph extraction method to replace the neighborhood subgraph.' The reference, however, is titled 'Differentially Private Graph Neural Networks for Link Prediction' and, by its title and venue, is about differential privacy, not path subgraph extraction. This appears to be a mismatch between the cited work and the summary. For a survey promising a 'rigorous comparative framework,' accurate method-to-reference mapping is load-bearing, and this error needs correction.
  3. [Section 5.1] Two method names in the knowledge-graph section are not expanded and appear garbled. 'FAGA' is introduced without an acronym definition and is cited to Li et al. (2024), whose title is 'Causal Subgraph Learning for Generalizable Inductive Relation Prediction' — there is no named FAGA model in that citation. Similarly, 'CEKF A' is not expanded; it presumably refers to the canonicalization-enhanced known-fact-aware framework of Wang et al. (2023c), but the spacing and letter are unexplained. Since the survey's utility depends on correctly linking method names to the original papers, these citation/terminology errors must be fixed.
  4. [Sections 2.2, 4.4] The GFormer category is conceptually vague and partly outside the stated 'GNN perspective.' Section 2.2 defines GFormer as 'a model that combines GNNs with the Transformer architecture' and cites the authors' own prior survey (Sun et al., 2023) rather than original graph-transformer works. Section 4.4 then discusses pure Transformer architectures (LPFormer, SIEG) as GFormer-based methods. If the survey's scope is GNN-based link prediction, non-GNN Transformers should either be excluded or explicitly argued to be part of the GNN family. At minimum, the definition must be sharpened and grounded in primary sources; as written, this category further undermines the taxonomy's exclusivity.
minor comments (4)
  1. [Throughout] The manuscript needs copyediting. Examples include 'Mutilayer' (Section 2.1), 'Recommendtion' (Figure 4), 'caculated' (Eq. 4 and surrounding text), 'exiting calssic' (Introduction), and inconsistent use of spaces around citations such as 'F AGA' and 'CEKF A.'
  2. [Table 1] A few table entries are inconsistent with the reference list. For example, 'PA' is listed as a model name for Subramonian et al. (2025), but the cited paper is titled 'Networked Inequality: Preferential Attachment Bias in Graph Neural Network Link Prediction' and the reference itself is dated ICML 2024; 'PA' is a concept, not a model name. The year discrepancy should also be reconciled.
  3. [Section 2.2] The GFormer definition cites Sun et al. (2023), a prior survey by the same authors, when introducing the combination of GNNs and Transformers. Citing the original architectural works (e.g., Graphormer, Dwivedi et al.'s graph transformer benchmark) would be more appropriate and would help establish that the category is independently grounded.
  4. [Section 6.2] The statement that GNNs 'have been shown to be incapable of differentiating node pairs that contain isomorphic nodes' is attributed to Zhang et al. (2021). That citation is appropriate, but the sentence should more precisely say 'node pairs whose endpoints are isomorphic in the enclosing subgraph' to avoid overgeneralizing; the current wording could be misread as a claim about all isomorphic nodes.

Circularity Check

0 steps flagged

No circular derivation: self-cited GFormer label and overlapping GAE/GCN categories affect taxonomy precision, but no result reduces to its inputs.

full rationale

This is a survey, so the relevant 'derivation chain' is the justification of the central organizing taxonomy, not a mathematical derivation. The paper's technique axis divides methods into GCN-, GAE-, GAT-, and GFormer-based backbones; this classification is an organizational choice, not a fitted or derived quantity. The GFormer category is introduced with a self-reference to the authors' prior survey ('GFormer is a model that combines GNNs with the Transformer architecture Sun et al. (2023)'), and the paper cites the authors' own LGAT/xGCN/LHGNN work in the preliminaries and Table 1. These are self-citations, but none is load-bearing in the sense of supplying an unverified premise on which a prediction rests: the Transformer self-attention mechanism is defined independently in Eqs. (7)-(10), and the surveyed GFormer-type methods (Graphormer, LPFormer, SIEG, etc.) are external works with published results. The GAE category is defined in Eq. (4) with a GCN encoder, so GAE-based and GCN-based categories overlap; Labeling Trick and NCNC are placed under GAE despite being labeling/pooling techniques. This is an internal-consistency weakness of the taxonomy, not a circular step: no parameter is fitted to one subset and then presented as a prediction on a related subset, and no conclusion follows by construction from the GFormer label. The central collecting/claiming function is supported by approximately eighty external references. Score 2 reflects the presence of self-citations at definitional points; no higher score is warranted because no derived or predicted result reduces to the survey's own prior work.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

As a survey, the central claim depends not on parameters or new entities but on the validity of the proposed taxonomy and the representativeness of the literature selection. The axioms listed are the structural premises underlying the survey's contribution.

axioms (3)
  • domain assumption GNN-based methods are the leading paradigm for link prediction (Section 1).
    The survey's premise for focusing on GNNs; not proven within the paper.
  • ad hoc to paper The four backbone categories GCN, GAE, GAT, and GFormer are mutually exclusive and jointly exhaustive for GNN-based link prediction methods (Section 3, Figure 4).
    The taxonomy is proposed by the authors without a formal derivation or validation that all methods fit these categories.
  • domain assumption The selected references in Table 1 are representative of the field (Section 3).
    No systematic search or inclusion criteria are provided; representativeness is assumed.

pith-pipeline@v1.3.0-alltime-deepseek · 24924 in / 8960 out tokens · 86738 ms · 2026-08-02T15:14:58.006262+00:00 · methodology

0 comments
read the original abstract

Graph Neural Networks (GNNs) have emerged as the leading paradigm for link prediction, enabling the inference of missing connections and the anticipation of potential future links. However, existing reviews lack systematic exploration specifically targeting underlying GNN architectures and diverse graph structures. To address this critical gap, this paper provides a comprehensive review of GNN-based link prediction from a novel and dedicated GNN perspective. We propose an innovative taxonomy that categorizes recent advancements based on techniques and applications. From a technique perspective, we focus on key GNN encoder architectures, including GCN-based, GAE-based, GAT-based, and GFormer-based methods, discussing their strengths and limitations. From an application perspective, we highlight prominent use cases of link prediction in knowledge graphs and recommendation systems, demonstrating their real-world impact. In addition, we examine the current challenges and discuss promising future directions.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

166 extracted references · 31 canonical work pages

  1. [4]

    In: Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing

    Berant J, Chou A, Frostig R, et al (2013) Semantic parsing on freebase from question-answer pairs. In: Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, pp 1533--1544

    2013

  2. [5]

    In: The Eleventh International Conference on Learning Representations, pp 1--12, ://openreview.net/forum?id=m1oqEOAozQU

    Chamberlain BP, Shirobokov S, Rossi E, et al (2023) Graph neural networks for link prediction with subgraph sketching. In: The Eleventh International Conference on Learning Representations, pp 1--12, ://openreview.net/forum?id=m1oqEOAozQU

    2023

  3. [8]

    Chen Z, Chen L, Villar S, et al (2020) Can graph neural networks count substructures? Advances in neural information processing systems 33:10383--10395

    2020

  4. [14]

    In: Proceedings of the First Learning on Graphs Conference, vol 198

    Dong K, Tian Y, Guo Z, et al (2022) Fakeedge: Alleviate dataset shift in link prediction. In: Proceedings of the First Learning on Graphs Conference, vol 198. PMLR, pp 56:1--56:19, ://openreview.net/forum?id=QDN0jSXuvtX

    2022

  5. [25]

    Expert Systems with Applications 264:125919

    Jiang F, Hu Q, Yang Z, et al (2025) A neighborhood rough sets-based ensemble method, with application to software fault prediction. Expert Systems with Applications 264:125919

    2025

  6. [26]

    Information Processing & Management 63(2):104483

    Jiang F, Yu X, Hu Q, et al (2026) An ensemble method using neighborhood granular combination entropy for software defect prediction. Information Processing & Management 63(2):104483

    2026

  7. [31]

    arXiv preprint arXiv:161107308

    Kipf TN, Welling M (2016) Variational graph auto-encoders. arXiv preprint arXiv:161107308

    2016

  8. [32]

    In: International Conference on Learning Representations, pp 1--14, ://openreview.net/forum?id=SJU4ayYgl

    Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: International Conference on Learning Representations, pp 1--14, ://openreview.net/forum?id=SJU4ayYgl

    2017

  9. [33]

    In: International Conference on Learning Representations, pp 1--21, ://openreview.net/forum?id=YicbFdNTTy

    Kolesnikov A, Dosovitskiy A, Weissenborn D, et al (2021) An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learning Representations, pp 1--21, ://openreview.net/forum?id=YicbFdNTTy

    2021

  10. [42]

    IEEE Transactions on Pattern Analysis and Machine Intelligence

    Liu J, Yang C, Lu Z, et al (2025) Graph foundation models: Concepts, opportunities and challenges. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2025

  11. [46]

    arXiv preprint arXiv:190711692

    Liu Y, Ott M, Goyal N, et al (2019) Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:190711692

    2019

  12. [52]

    In: Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, IJCAI '23, doi:10.24963/ijcai.2023/248, ://doi.org/10.24963/ijcai.2023/248

    Luo S, Li H, Huang J (2023 b ) Dynamic group link prediction in continuous-time interaction network. In: Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, IJCAI '23, doi:10.24963/ijcai.2023/248, ://doi.org/10.24963/ijcai.2023/248

    work page doi:10.24963/ijcai.2023/248 2023

  13. [54]

    Advances in Neural Information Processing Systems 36:79594--79612

    Luo Z, Huang H, Lian J, et al (2023 d ) Cross-links matter for link prediction: rethinking the debiased gnn from a data perspective. Advances in Neural Information Processing Systems 36:79594--79612

    2023

  14. [57]

    In: Proceedings of the 34th ACM International Conference on Information and Knowledge Management

    Ma W, Wang Y, Wang X, et al (2025) Reconsidering the performance of gae in link prediction. In: Proceedings of the 34th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, New York, NY, USA, CIKM '25, p 2052–2062, doi:10.1145/3746252.3761108, ://doi.org/10.1145/3746252.3761108

    arXiv 2025

  15. [65]

    In: International Conference on Learning Representations, pp 1--18, ://openreview.net/forum?id=CCu6RcUMwK0

    Pan L, Shi C, Dokmani \'c I (2022) Neural link prediction with walk pooling. In: International Conference on Learning Representations, pp 1--18, ://openreview.net/forum?id=CCu6RcUMwK0

    2022

  16. [72]

    Social Network Analysis and Mining 10(1):67

    Rehman SU, Asghar S (2020) Online social network trend discovery using frequent subgraph mining. Social Network Analysis and Mining 10(1):67

    2020

  17. [73]

    In: Seventh International Conference on Digital Information Management (ICDIM 2012), IEEE, pp 88--92

    Rehman SU, Khan AU, Fong S (2012) Graph mining: A survey of graph mining techniques. In: Seventh International Conference on Digital Information Management (ICDIM 2012), IEEE, pp 88--92

    2012

  18. [74]

    Mathematical problems in engineering 2014(1):869198

    Rehman SU, Asghar S, Zhuang Y, et al (2014) Performance evaluation of frequent subgraph discovery techniques. Mathematical problems in engineering 2014(1):869198

    2014

  19. [75]

    In: Proceedings of the 2018 10th international conference on machine learning and computing, pp 257--262

    Rehman SU, Asghar S, Fong S (2018) An efficient ranking scheme for frequent subgraph patterns. In: Proceedings of the 2018 10th international conference on machine learning and computing, pp 257--262

    2018

  20. [76]

    Research Updates in Mathematics and Computer Science 4:33--63

    Rehman SU, Khalil MI, Kundi M, et al (2024) A study on frequent subgraph mining approaches: Challenges and future directions. Research Updates in Mathematics and Computer Science 4:33--63

    2024

  21. [78]

    EAI Endorsed Transactions on Industrial Networks and Intelligent Systems 7(23)

    Samad A, Qadir M, Nawaz I, et al (2020) A comprehensive survey of link prediction techniques for social network. EAI Endorsed Transactions on Industrial Networks and Intelligent Systems 7(23). doi:10.4108/eai.13-7-2018.163988

    arXiv 2020

  22. [81]

    In: Companion Proceedings of the ACM Web Conference 2024, pp 1264--1267

    Shi C, Yang C, Fang Y, et al (2024 a ) Lecture-style tutorial: Towards graph foundation models. In: Companion Proceedings of the ACM Web Conference 2024, pp 1264--1267

    2024

  23. [84]

    In: International Conference on Learning Representations, pp 1--17

    Skarding J, Hellmich M, Gabrys B, et al (2022) Benchmarking graph neural networks on dynamic link prediction. In: International Conference on Learning Representations, pp 1--17

    2022

  24. [88]

    In: Proceedings of the 41st International Conference on Machine Learning , ICML '24, vol 235

    Subramonian A, Sagun L, Sun Y (2025) Networked inequality: preferential attachment bias in graph neural network link prediction. In: Proceedings of the 41st International Conference on Machine Learning , ICML '24, vol 235. JMLR.org, Vienna, Austria, pp 46891--46925

    2025

  25. [92]

    arXiv preprint arXiv:241001802

    Tola A, Myrick J, Coskunuzer B (2024) Proxi: Challenging the gnns for link prediction. arXiv preprint arXiv:241001802

    2024

  26. [93]

    In: Proceedings of the 31st International Conference on Neural Information Processing Systems

    Vaswani A, Shazeer N, Parmar N, et al (2017) Attention is all you need. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. Curran Associates Inc., Red Hook, NY, USA, NIPS'17, pp 6000--6010

    2017

  27. [94]

    In: International Conference on Learning Representations, pp 1--12, ://openreview.net/forum?id=rJXMpikCZ

    Veličković P, Cucurull G, Casanova A, et al (2018) Graph attention networks. In: International Conference on Learning Representations, pp 1--12, ://openreview.net/forum?id=rJXMpikCZ

    2018

  28. [96]

    In: International Conference on Learning Representations, pp 1--24, ://openreview.net/forum?id=e95i1IHcWj

    Wang H, Yin H, Zhang M, et al (2022) Equivariant and stable positional encoding for more powerful graph neural networks. In: International Conference on Learning Representations, pp 1--24, ://openreview.net/forum?id=e95i1IHcWj

    2022

  29. [98]

    In: The Twelfth International Conference on Learning Representations, pp 1--17

    Wang X, Yang H, Zhang M (2024 a ) Neural common neighbor with completion for link prediction. In: The Twelfth International Conference on Learning Representations, pp 1--17

    2024

  30. [101]

    In: The Twelfth International Conference on Learning Representations, pp 1--39

    Wang Y, Zhao T, Zhao Y, et al (2024 c ) A topological perspective on demystifying gnn-based link prediction performance. In: The Twelfth International Conference on Learning Representations, pp 1--39

    2024

  31. [110]

    Curran Associates, Inc., pp 28877--28888

    Ying C, Cai T, Luo S, et al (2021) Do transformers really perform badly for graph representation ? In: Advances in Neural Information Processing Systems , vol 34. Curran Associates, Inc., pp 28877--28888

    2021

  32. [116]

    arXiv preprint arXiv:240607926

    Zhang X, Wang Y, Wang X, et al (2024 b ) Efficient neural common neighbor for temporal graph link prediction. arXiv preprint arXiv:240607926

    2024

  33. [124]

    arXiv preprint arXiv:220801820

    Zhou S, Guo Z, Aggarwal C, et al (2022) Link prediction on heterophilic graphs via disentangled representation learning. arXiv preprint arXiv:220801820

    2022

  34. [126]

    Advances in Neural Information Processing Systems 36:21368--21414

    Zhou Z, Yao J, Liu J, et al (2023) Combating bilateral edge noise for robust link prediction. Advances in Neural Information Processing Systems 36:21368--21414

    2023

  35. [127]

    Semantic Parsing on Freebase from Question-Answer Pairs

    Berant, Jonathan and Chou, Andrew and Frostig, Roy and Liang, Percy. Semantic Parsing on Freebase from Question-Answer Pairs. Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing. 2013

    2013

  36. [128]

    Advances in Neural Information Processing Systems , volume=

    Cross-links matter for link prediction: rethinking the debiased GNN from a data perspective , author=. Advances in Neural Information Processing Systems , volume=

  37. [129]

    and Shah, Neil and Zhao, Tong , month = jul, year =

    Guo, Zhichun and Shiao, William and Zhang, Shichang and Liu, Yozen and Chawla, Nitesh V. and Shah, Neil and Zhao, Tong , month = jul, year =. Linkless link prediction via relational distillation , volume =. Proceedings of the 40th

  38. [130]

    Do Transformers Really Perform Badly for Graph Representation ? , volume =

    Ying, Chengxuan and Cai, Tianle and Luo, Shengjie and Zheng, Shuxin and Ke, Guolin and He, Di and Shen, Yanming and Liu, Tie-Yan , year =. Do Transformers Really Perform Badly for Graph Representation ? , volume =. Advances in

  39. [131]

    Networked inequality: preferential attachment bias in graph neural network link prediction , volume =

    Subramonian, Arjun and Sagun, Levent and Sun, Yizhou , month = jan, year =. Networked inequality: preferential attachment bias in graph neural network link prediction , volume =. Proceedings of the 41st

  40. [132]

    and Kaiser,

    Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N. and Kaiser,. Attention is all you need , year =. Proceedings of the 31st International Conference on Neural Information Processing Systems , pages =

  41. [133]

    Advances in Neural Information Processing Systems , author =

    Combating Bilateral Edge Noise for Robust Link Prediction , volume =. Advances in Neural Information Processing Systems , author =. 2023 , pages =

    2023

  42. [134]

    The Twelfth International Conference on Learning Representations , year=

    A topological perspective on demystifying gnn-based link prediction performance , author=. The Twelfth International Conference on Learning Representations , year=

  43. [135]

    arXiv preprint arXiv:2410.01802 , year=

    PROXI: Challenging the GNNs for Link Prediction , author=. arXiv preprint arXiv:2410.01802 , year=

    arXiv

  44. [136]

    IEEE Transactions on Knowledge and Data Engineering , volume=

    GTRL: An entity group-aware temporal knowledge graph representation learning method , author=. IEEE Transactions on Knowledge and Data Engineering , volume=. 2023 , publisher=

    2023

  45. [137]

    arXiv preprint arXiv:1907.11692 , year=

    Roberta: A robustly optimized bert pretraining approach , author=. arXiv preprint arXiv:1907.11692 , year=

    Pith/arXiv arXiv 1907

  46. [138]

    The Twelfth International Conference on Learning Representations , year =

    Neural Common Neighbor with Completion for Link Prediction , author =. The Twelfth International Conference on Learning Representations , year =

  47. [139]

    arXiv preprint arXiv:2208.01820 , year=

    Link prediction on heterophilic graphs via disentangled representation learning , author=. arXiv preprint arXiv:2208.01820 , year=

    Pith/arXiv arXiv

  48. [140]

    arXiv preprint arXiv:2406.07926 , year=

    Efficient Neural Common Neighbor for Temporal Graph Link Prediction , author=. arXiv preprint arXiv:2406.07926 , year=

    arXiv

  49. [141]

    Advances in neural information processing systems , volume=

    Can graph neural networks count substructures? , author=. Advances in neural information processing systems , volume=

  50. [142]

    arXiv preprint arXiv:1611.07308 , year=

    Variational graph auto-encoders , author=. arXiv preprint arXiv:1611.07308 , year=

    Pith/arXiv arXiv

  51. [143]

    International Conference on Learning Representations , month = oct, year =

    Benchmarking Graph Neural Networks on Dynamic Link Prediction , author =. International Conference on Learning Representations , month = oct, year =

  52. [144]

    International Conference on Learning Representations , year=

    Semi-Supervised Classification with Graph Convolutional Networks , author=. International Conference on Learning Representations , year=

  53. [145]

    International Conference on Learning Representations , year=

    Graph Attention Networks , author=. International Conference on Learning Representations , year=

  54. [146]

    Physical Review E , volume =

    Clustering and preferential attachment in growing networks , author =. Physical Review E , volume =. 2001 , month =. doi:10.1103/PhysRevE.64.025102

    doi:10.1103/physreve.64.025102 2001

  55. [147]

    Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , pages =

    Perozzi, Bryan and Al-Rfou, Rami and Skiena, Steven , title =. Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , pages =. 2014 , publisher =. doi:10.1145/2623330.2623732

    arXiv 2014

  56. [148]

    Expert Systems with Applications , volume=

    Link prediction by deep non-negative matrix factorization , author=. Expert Systems with Applications , volume=. 2022 , publisher=. doi:10.1016/j.eswa.2021.115991

    arXiv 2022

  57. [149]

    Journal of Big Data , volume=

    A review of graph neural networks: concepts, architectures, techniques, challenges, datasets, applications, and future directions , author=. Journal of Big Data , volume=. 2024 , publisher=. doi:10.1186/s40537-023-00876-4

    doi:10.1186/s40537-023-00876-4 2024

  58. [150]

    EAI Endorsed Transactions on Industrial Networks and Intelligent Systems , volume=

    Abdul Samad and Mamoona Qadir and Ishrat Nawaz and Muhammad Arshad Islam and Muhammad Aleem , title=. EAI Endorsed Transactions on Industrial Networks and Intelligent Systems , volume=. 2020 , month=

    2020

  59. [151]

    Journal of Machine Learning Research , volume=

    Benchmarking graph neural networks , author=. Journal of Machine Learning Research , volume=. doi:10.5555/3648699.3648742

    arXiv

  60. [152]

    ACM Computing Surveys (CSUR) , volume=

    A survey on embedding dynamic graphs , author=. ACM Computing Surveys (CSUR) , volume=. 2021 , publisher=. doi:10.1145/3483595

    work page doi:10.1145/3483595 2021

  61. [153]

    Knowledge-Based Systems , volume=

    Community preserving adaptive graph convolutional networks for link prediction in attributed networks , author=. Knowledge-Based Systems , volume=. 2023 , publisher=. doi:10.1016/j.knosys.2023.110589

    arXiv 2023

  62. [154]

    Information Processing & Management , author =

    Edge contrastive learning for link prediction , volume =. Information Processing & Management , author =. 2024 , pages =. doi:10.1016/j.ipm.2024.103847

    arXiv 2024

  63. [155]

    Advances in Neural Information Processing Systems , volume=

    Neo-gnns: Neighborhood overlap-aware graph neural networks for link prediction , author=. Advances in Neural Information Processing Systems , volume=. doi:10.5555/3540261.3541309

    arXiv

  64. [156]

    Computing , volume=

    LGAT: a light graph attention network focusing on message passing for semi-supervised node classification , author=. Computing , volume=. doi:10.1007/s00607-024-01261-6

    work page doi:10.1007/s00607-024-01261-6

  65. [157]

    Advances in Neural Information Processing Systems , volume=

    Labeling trick: A theory of using graph neural networks for multi-node representation learning , author=. Advances in Neural Information Processing Systems , volume=. doi:10.5555/3540261.3540954

    arXiv

  66. [158]

    Proceedings of the 34th ACM International Conference on Information and Knowledge Management , pages =

    Ma, Weishuo and Wang, Yanbo and Wang, Xiyuan and Zhang, Muhan , title =. Proceedings of the 34th ACM International Conference on Information and Knowledge Management , pages =. 2025 , isbn =

    2025

  67. [159]

    International Conference on Learning Representations , year=

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale , author=. International Conference on Learning Representations , year=

  68. [160]

    Sun, Chengcheng and Li, Chenhao and Lin, Xiang and Zheng, Tianji and Meng, Fanrong and Rui, Xiaobin and Wang, Zhixiao , title =. Artif. Intell. Rev. , month = aug, pages =. 2023 , issue_date =. doi:10.1007/s10462-023-10577-2

    work page doi:10.1007/s10462-023-10577-2 2023

  69. [161]

    Bert: Pre-training of deep bidirectional transformers for language understanding , author=. Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers) , pages=. doi:10.18653/v1/N19-1423

    doi:10.18653/v1/n19-1423 2019

  70. [162]

    Information Processing & Management , author =

    Knowledge graph embedding based on dynamic adaptive atrous convolution and attention mechanism for link prediction , volume =. Information Processing & Management , author =. 2024 , pages =. doi:10.1016/j.ipm.2024.103642

    arXiv 2024

  71. [163]

    Applied Soft Computing , volume=

    A double attention graph network for link prediction on temporal graph , author=. Applied Soft Computing , volume=. 2023 , publisher=. doi:10.1016/j.asoc.2023.110059

    arXiv 2023

  72. [164]

    Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence , articleno =

    Luo, Shijie and Li, He and Huang, Jianbin , title =. Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence , articleno =. 2023 , isbn =

    2023

  73. [165]

    Expert Systems with Applications , volume=

    Link prediction in heterogeneous networks based on metapath projection and aggregation , author=. Expert Systems with Applications , volume=. 2023 , publisher=. doi:10.1016/j.eswa.2023.120325

    arXiv 2023

  74. [166]

    FakeEdge: Alleviate Dataset Shift in Link Prediction , volume =

    Dong, Kaiwen and Tian, Yijun and Guo, Zhichun and Yang, Yang and Chawla, Nitesh , month = dec, year =. FakeEdge: Alleviate Dataset Shift in Link Prediction , volume =. Proceedings of the First Learning on Graphs Conference , publisher =

  75. [167]

    International Conference on Learning Representations , year=

    Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networks , author=. International Conference on Learning Representations , year=

  76. [168]

    Pattern Recognition , volume=

    Line graph contrastive learning for link prediction , author=. Pattern Recognition , volume=. 2023 , publisher=. doi:10.1016/j.patcog.2023.109537

    arXiv 2023

  77. [169]

    International Conference on Learning Representations , year=

    Neural Link Prediction with Walk Pooling , author=. International Conference on Learning Representations , year=

  78. [170]

    The Eleventh International Conference on Learning Representations , year=

    Graph Neural Networks for Link Prediction with Subgraph Sketching , author=. The Eleventh International Conference on Learning Representations , year=

  79. [171]

    Proceedings of the 32nd International Conference on Neural Information Processing Systems , pages =

    Zhang, Muhan and Chen, Yixin , title =. Proceedings of the 32nd International Conference on Neural Information Processing Systems , pages =. 2018 , publisher =. doi:10.5555/3327345.3327423

    arXiv 2018

  80. [172]

    Technological Forecasting and Social Change , volume=

    Two-stage technology opportunity discovery for firm-level decision making: GCN-based link-prediction approach , author=. Technological Forecasting and Social Change , volume=. 2022 , publisher=. doi:10.1016/j.techfore.2022.121934

    arXiv 2022

Showing first 80 references.