Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A cluster layer on top of the item graph improves graph contrastive recommendation.

desk verdict An incremental GCL extension with a plausible but under-validated clustering mechanism; the gains over XSimGCL are real but tiny, and the paper needs robustness analysis before the hierarchy claim can be trusted. read the letter →

arxiv 2505.19020 v1 pith:B6A36N5C submitted 2025-05-25 cs.IR cs.AI

Jiawei Xue , Zhen Yang , Haitao Lin , Ziji Zhang , Luzhu Wang , Yikun Gu , Yao Xu , Xin Li This is my paper
classification cs.IRcs.AI
keywords hierarchicalgraphcontrastivelearningneuralnetworksrecommendationsystemsitemclusteringt-SNEuser-itembipartiterepresentation
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

Hierarchical item structure, discovered from learned embeddings rather than item metadata, is the missing signal this paper adds to graph contrastive recommendation. HGCL pre-trains a GCL model on the user-item graph, projects the item embeddings into two dimensions with t-SNE, and divides that plane into $\rho$ radial and $\theta$ angular sectors; each sector becomes a cluster node in a second user-to-cluster graph. Fine-tuning on both graphs lets cluster-level collaborative signal enter the final item score, and the paper reports the highest Recall@20 and NDCG@20 among seven models on Yelp2018, Amazon-Kindle, and Alibaba-iFashion. The comparison to XSimGCL, the method HGCL extends, is the evidence offered that the hierarchy itself, not the extra parameters, drives the gain.

What carries the argument

The load-bearing mechanism is the two-hierarchy user-item bipartite graph built by polar-sector clustering in a t-SNE plane. Pre-trained item embeddings are compressed to two dimensions with t-SNE; the plane is partitioned into $\rho$ radial bands and $\theta$ angular sectors, and every item whose projected point falls in a sector joins that cluster. A separate graph links each user to the clusters containing items the user interacted with, and the two graphs are convolved with the same LightGCN-style operator. The final connecting strength $\hat{y}_{ij}$ uses a weighted sum of the item embedding and all cluster embeddings containing the item, so the cluster representations act as learnable category prototypes that refine the item's place in ranking.

What would settle it

Run HGCL on Yelp2018 with multiple t-SNE random seeds and with different choices of the polar origin (coordinate origin, centroid, median), keeping $\rho$ and $\theta$ fixed. If the Recall@20 advantage over XSimGCL (0.0736 versus 0.0726) disappears or changes sign under any of those variations, the claim that hierarchical item structure causes the gain is not supported. A complementary check is a permutation test: assign items to sectors at random with the same cluster sizes and same number of clusters; if random sectors match HGCL's gain, the specific t-SNE-derived clusters are not doing the work.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a two-level user-item graph, with original items plus item clusters, yields better recommendations than a flat user-item graph under the same GCL machinery. Item clusters are formed deterministically by clustering t-SNE-transformed item embeddings into polar sectors, so the hierarchy is representation-driven and needs no category labels. The final score for a user and item is the dot product of the user embedding with the item embedding plus the embedding of the item's cluster, which channels coarse-grained signals into ranking. Across three benchmark datasets the paper reports consistent, though small, gains over XSimGCL, for example Recall@20 of 0.0736 versus 0.0726 on Yelp2018, and takes these as evidence that hierarchical item structures matter for graph contrastive recommendation.

Load-bearing premise

The entire gain rests on the assumption that cutting a t-SNE projection of item embeddings into fixed polar sectors produces item clusters that capture similarities that actually help recommendation; the paper does not show that the result is stable under changes to the t-SNE random seed, the polar origin, or the sector boundaries.

Editorial extensions

If this is right

  • Any GCL recommender can be extended by adding a cluster-level graph on top of the item graph, since the hierarchy is derived from learned embeddings and requires no item attributes or external knowledge.
  • Recommendations become multi-resolution: a user can be matched to an item directly or through a cluster, which should let items with few interactions inherit collaborative signal from the cluster.
  • The number of radial and angular divisions is dataset-dependent; the sensitivity results show that too many sectors hurt performance, so practitioners need to tune $\rho$ and $\theta$ for each application.
  • The reported shift in connection-strength distributions after fine-tuning indicates the cluster layer mainly sharpens the separation between positive and negative user-item pairs.
  • Because HGCL inherits the pre-trained GCL weights from XSimGCL, its gains are additive to whatever uniformity and contrastive improvements the base method already provides.

Reading between the lines

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

  • An untested consequence of the paper's reasoning is that cold-start and long-tail items should benefit disproportionately from the cluster layer, since they are the items with the least direct collaborative evidence; breaking down the reported gains by item popularity would test this directly.
  • The polar-sector partition is one of many possible clusterings of the t-SNE plane; a natural extension the paper does not explore is whether soft membership weights, or clustering done directly in the original embedding space, preserve or enlarge the gain.
  • Because sector boundaries are arbitrary, items near a boundary are assigned to one cluster only; in practice, overlapping or fuzzy sectors could smooth the boundary effect and may interact with the choice of $\rho$ and $\theta$.
  • If the gain is due to cluster prototypes acting as regularizers, similar or larger gains might be obtainable by adding a cluster-consistency contrastive loss on the item embeddings themselves, a modification that would avoid t-SNE and its stochasticity.
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 / 6 minor

Summary. The paper proposes HGCL, a graph contrastive learning method for user-item recommendation that extends XSimGCL. The method first pre-trains user/item representations with cross-layer contrastive learning, then projects item embeddings to two dimensions with t-SNE and partitions the plane into radial and angular sectors to form item clusters. These clusters are used to build a user-clustered item graph, and the model is fine-tuned jointly on the original and clustered graphs, with the final score for a user-item pair augmented by the item's cluster embedding (Eq. 7). Experiments on Yelp2018, Amazon-Kindle, and Alibaba-iFashion report Recall@20 and NDCG@20 for seven methods and claim that HGCL outperforms all baselines, including XSimGCL, thereby demonstrating the value of hierarchical item structures.

Significance. If the empirical claims were robust, the paper would make a modest but useful contribution: a simple, model-agnostic way to inject coarse item-cluster information into an existing GCL recommender, with a clear two-stage pipeline and sensitivity analyses for the cluster-division and perplexity hyperparameters. The paper also openly acknowledges that scalability to large industrial datasets is not demonstrated. However, the central quantitative claim is supported only by point estimates with no variance information, and the cluster construction is under-specified because the polar origin is not defined and the t-SNE randomness is not controlled. The reported gains over XSimGCL are small, and the paper does not provide a validation protocol for the dataset-specific hyperparameters. No code is released, so reproducibility currently depends on a written specification that is incomplete. The idea is plausible but the evidence is not yet sufficient to establish the headline attribution of the gains to hierarchical item structures.

major comments (4)
  1. [Sec. 4.2, Fig. 1] The radial/angular sector clustering is not a well-defined function of the data because the polar origin used to divide the t-SNE plane is never specified. Translating or rotating E_tsne changes every cluster boundary, and the paper does not state how the origin is selected or whether the coordinates are aligned before sectoring. Since the cluster memberships w_jk enter the final score in Eq. (7), the method is not reproducible as written; please specify the origin (e.g., the centroid of E_tsne) and any normalization or alignment step.
  2. [Sec. 4.2, Eq. (6); Sec. 5.3, Fig. 3] The t-SNE projection is stochastic, but no random seed, number of runs, or aggregation across runs is reported. The sensitivity study in Fig. 3 varies rho and theta but never varies the t-SNE run, so it cannot rule out the possibility that the reported gains over XSimGCL are artifacts of one particular t-SNE draw. Please report the mean and standard deviation over several t-SNE runs, or fix and disclose the seed and verify that the qualitative conclusions are stable across runs.
  3. [Sec. 5.2, Table 2; Sec. 5.3, Fig. 5] The central claim of superior performance and the attribution of that performance to hierarchical item structures rest on point estimates without error bars or significance tests. The improvements over XSimGCL are small, for example Yelp2018 Recall@20 0.0736 vs 0.0726, Amazon-Kindle 0.2117 vs 0.2061, and Alibaba-iFashion 0.1194 vs 0.1189, with similarly small NDCG differences. The connecting-strength analysis in Fig. 5 is descriptive only and reports no uncertainty. Please provide multiple seeds, standard deviations, and, where appropriate, paired significance tests to support the Q2 conclusion.
  4. [Sec. 5.3, Figs. 3-4; Table 1] The hyperparameters rho, theta, and perplexity are tuned per dataset, and the values in Table 1 appear to be the ones that maximize test Recall@20 according to Figs. 3 and 4, but no validation split or model-selection protocol is described. If test performance is used to choose these hyperparameters, the reported results are selected on the test set and the gains relative to baselines are inflated. Please specify a validation procedure, or report results for fixed default settings in addition to the tuned settings.
minor comments (6)
  1. [Sec. 4.2, Eq. (6)] The formula for p_xy is not the standard t-SNE conditional probability with perplexity; the notation 'k,l within the neighborhood' is vague, and the global normalization shown does not match the usual definition. Please align Eq. (6) with the standard t-SNE formulation or explain the simplified version used here.
  2. [Sec. 4.2] The term 'hierarchical item structures' suggests multiple resolutions of item similarity, but the construction adds only a single level of item clusters. The abstract's claim of structures 'across varying resolutions' is stronger than what is actually implemented; consider clarifying that the hierarchy has two levels.
  3. [Sec. 5.1.2] There is a typo: 'betwen' should be 'between'.
  4. [Sec. 4.2] The word 'persevering' appears where 'preserving' is intended.
  5. [Table 2] The footnote states that BUIR and DNN+SSL results are sourced from [38]; please state explicitly whether those numbers were obtained under the same data splits, evaluation protocols, and hyperparameter settings as the other baselines.
  6. [Figs. 3-4] If multiple runs are performed after addressing the stochasticity concerns, the sensitivity figures would be much more informative with error bars or shaded confidence bands.

Circularity Check

2 steps flagged · score 5.0 of 10

Cluster geometry and t-SNE perplexity are selected on the same test metrics used to claim that hierarchical structures help, so the Q2 'prediction' is partly a fitted outcome; no other circular steps found.

  1. fitted input called prediction [Section 5.3 (Sensitivity study), Table 1, Fig. 3; coupled with Sec. 5.2 Q2 answer]
    "The results show that setting θ= 2 on the Yelp2018 and Amazon-Kindle datasets yield optimal performance, whereas θ= 8, ρ= 4 leads to highest accuracy on the Alibaba-iFashion dataset."

    At least for Alibaba-iFashion the reported optimum (ρ=4, θ=8) is exactly the configuration used in Table 1, so the claimed gain is a directly fitted outcome; for all datasets the sensitivity study chooses ρ/θ by the same Recall@20 on which HGCL is later compared. Sec. 5.2 states that the HGCL-vs-XSimGCL gap 'further substantiates the beneficial contribution of hierarchical item structures.' Selecting the cluster-defining geometry on the evaluation metric and then presenting that metric as evidence makes the Q2 answer a fitted input rather than an independent prediction; the advantage is partly a selection artifact.

  2. fitted input called prediction [Section 5.3 (Sensitivity study), Fig. 4]
    "the perplexity setting of 30 yields the highest Recall@20 and NDCG@20 scores among the five tested values on the Yelp2018 and Alibaba-iFashion datasets."

    Perplexity controls the t-SNE neighborhood size from which the item clusters are derived. Choosing it by the test-set Recall@20 and NDCG@20 scores, and then using those same test scores to demonstrate that the hierarchical clustering helps, folds the evaluation metric into the cluster construction. The statement that perplexity 30 is best is a selection result on the headline metrics, not a parameter-free prediction of hierarchical structure benefit.

full rationale

The core pipeline is not circular: item clusters are derived from pre-trained embeddings via t-SNE (Sec. 4.2), not from the evaluation labels, and HGCL is built on the external XSimGCL baseline with an added user-cluster graph and fine-tuning objective (Sec. 4.3). The self-citation [31] is used only for problem formulation and is not load-bearing. No uniqueness theorem or ansatz is smuggled in by self-citation, and no known result is renamed. The circular element is confined to the experimental handling of the hierarchy claim: the cluster-defining hyperparameters ρ, θ, and t-SNE perplexity are selected using the same Recall@20/NDCG@20 metrics that are later reported as evidence that hierarchical item structures contribute (Q2). That makes the Q2 'prediction' partly a fitted outcome rather than a fully independent validation. The overall superiority claim over six baselines retains independent content because the benchmark comparisons and the pretraining/fine-tuning architecture are external to the fitted cluster parameters, so the paper is only partially circular, not a derivation equivalent to its inputs.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central claim rests on the assumption that t-SNE plus polar-coordinate clustering creates meaningful item categories, and on two per-dataset hyperparameters (ρ and θ) plus a tuned perplexity. No independent evidence is provided that the clusters are stable or semantically valid.

free parameters (3)
  • ρ (radial divisions) = Yelp2018: 8; Amazon-Kindle: 1; Alibaba-iFashion: 4
    Number of radial sectors in the polar clustering; tuned per dataset to maximize Recall@20 (Fig. 3).
  • θ (angular divisions) = Yelp2018: 4; Amazon-Kindle: 4; Alibaba-iFashion: 8
    Number of angular sectors; tuned per dataset to maximize Recall@20 (Fig. 3).
  • t-SNE perplexity = 30 (selected from five tested values)
    Controls the neighborhood size in t-SNE; chosen based on best Recall and NDCG in Fig. 4.
assumptions (4)
  • domain assumption User-item interactions form a bipartite graph and BPR loss is appropriate for ranking.
    Invoked in Section 3.1 and used throughout the method.
  • domain assumption t-SNE projection preserves local neighborhood relationships that reflect item similarity relevant to recommendation.
    Stated in Section 4.2 as the reason for choosing t-SNE over PCA.
  • ad hoc to paper Dividing the t-SNE plane into radial and angular sectors creates meaningful item clusters.
    The sector division is introduced in Section 4.2 with no justification beyond simplicity and interpretability.
  • domain assumption The cluster-level graph provides useful collaborative signal for items with few interactions.
    This is the core hypothesis behind the fine-tuning stage, stated in Section 1 but not independently tested.
invented entities (1)
  • Clustered item nodes (C) and the user-clustered item graph
    purpose: To inject coarse-grained item similarity into the final scoring function (Eq. 7).
    The clusters are constructed from the same training data used for evaluation, so the reported performance gains are not independent confirmation; no external validation of the clusters is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation." pith.science (2026). https://pith.science/paper/B6A36N5C

@misc{pith2026250519020,
  author       = {Pith},
  title        = {Pith review of: HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B6A36N5C}},
  note         = {Machine review of arXiv:2505.19020}
}
read the original abstract

Graph Contrastive Learning (GCL), which fuses graph neural networks with contrastive learning, has evolved as a pivotal tool in user-item recommendations. While promising, existing GCL methods often lack explicit modeling of hierarchical item structures, which represent item similarities across varying resolutions. Such hierarchical item structures are ubiquitous in various items (e.g., online products and local businesses), and reflect their inherent organizational properties that serve as critical signals for enhancing recommendation accuracy. In this paper, we propose Hierarchical Graph Contrastive Learning (HGCL), a novel GCL method that incorporates hierarchical item structures for user-item recommendations. First, HGCL pre-trains a GCL module using cross-layer contrastive learning to obtain user and item representations. Second, HGCL employs a representation compression and clustering method to construct a two-hierarchy user-item bipartite graph. Ultimately, HGCL fine-tunes user and item representations by learning on the hierarchical graph, and then provides recommendations based on user-item interaction scores. Experiments on three widely adopted benchmark datasets ranging from 70K to 382K nodes confirm the superior performance of HGCL over existing baseline models, highlighting the contribution of hierarchical item structures in enhancing GCL methods for recommendation tasks.

Figures

Figures reproduced from arXiv: 2505.19020 by the authors.

Figure 1
Figure 1. The framework of HGCL. Initially, HGCL generates user and item representations using [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Learning curves of different models on the three datasets. [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Recall@20 scores from HGCL with different values of hyperparameters [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Recall@20 and NDCG@20 scores from HGCL with different values of perplexity in t-SNE. [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: The connecting strength of positive and negative user-item pairs after pre-training ( [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. GR-LLMs: Recent Advances in Generative Recommendation Based on Large Language Models

    cs.IR 2025-07 unverdicted novelty 3.0 of 10

    A survey of LLM-based generative recommendation systems, covering application settings, training pipelines, industrial deployment challenges, and future directions.

Reference graph

Works this paper leans on

43 extracted references · 33 canonical work pages · cited by 1 Pith paper

  1. [1]

    OmniSage: Large Scale, Multi-Entity Heterogeneous Graph Representation Learning

    Anirudhan Badrinath, Alex Yang, Kousik Rajesh, Prabhat Agarwal, Jaewon Yang, Haoyu Chen, Jiajing Xu, and Charles Rosenberg. OmniSage: Large scale, multi-entity heterogeneous graph representation learning.ArXiv preprint arXiv:2504.17811, 2025

    work page Pith review arXiv 2025

  2. [2]

    LiGNN: Graph neural networks at LinkedIn

    Fedor Borisyuk, Shihai He, Yunbo Ouyang, Morteza Ramezani, Peng Du, Xiaochen Hou, Chengming Jiang, Nitin Pasumarthy, Priya Bannur, Birjodh Tiwana, et al. LiGNN: Graph neural networks at LinkedIn. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 4793–4803, 2024

    work page 2024

  3. [3]

    LightGCL: Simple yet effective graph contrastive learning for recommendation

    Xuheng Cai, Chao Huang, Lianghao Xia, and Xubin Ren. LightGCL: Simple yet effective graph contrastive learning for recommendation. InThe Eleventh International Conference on Learning Representations, 2023

    work page 2023

  4. [4]

    Macro graph neural networks for online billion-scale recommender systems

    Hao Chen, Yuanchen Bei, Qijie Shen, Yue Xu, Sheng Zhou, Wenbing Huang, Feiran Huang, Senzhang Wang, and Xiao Huang. Macro graph neural networks for online billion-scale recommender systems. InProceedings of the ACM Web Conference 2024, pages 3598–3608, 2024

    work page 2024

  5. [5]

    Leveraging contrastive learning for enhanced node representations in tokenized graph transformers.Advances in Neural Information Processing Systems, 37:85824–85845, 2024

    Jinsong Chen, Hanpeng Liu, John Hopcroft, and Kun He. Leveraging contrastive learning for enhanced node representations in tokenized graph transformers.Advances in Neural Information Processing Systems, 37:85824–85845, 2024

    work page 2024

  6. [6]

    Heteroge- neous graph contrastive learning for recommendation

    Mengru Chen, Chao Huang, Lianghao Xia, Wei Wei, Yong Xu, and Ronghua Luo. Heteroge- neous graph contrastive learning for recommendation. InProceedings of the Sixteenth ACM International Conference on Web Search and Data Mining, pages 544–552, 2023

    work page 2023

  7. [7]

    A simple framework for contrastive learning of visual representations

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual representations. InInternational Conference on Machine Learning, pages 1597–1607. PMLR, 2020

    2020

  8. [8]

    Understanding the difficulty of training deep feedforward neural networks

    Xavier Glorot and Yoshua Bengio. Understanding the difficulty of training deep feedforward neural networks. InProceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pages 249–256. JMLR Workshop and Conference Proceedings, 2010

    2010

Show all 43 references
  1. [9]

    Architecture matters: Uncovering implicit mechanisms in graph contrastive learning.Advances in Neural Information Processing Systems, 36:28585–28610, 2023

    Xiaojun Guo, Yifei Wang, Zeming Wei, and Yisen Wang. Architecture matters: Uncovering implicit mechanisms in graph contrastive learning.Advances in Neural Information Processing Systems, 36:28585–28610, 2023

    2023

  2. [10]

    Exploitation of a latent mechanism in graph contrastive learning: Representation scattering

    Dongxiao He, Lianze Shan, Jitao Zhao, Hengrui Zhang, Zhen Wang, and Weixiong Zhang. Exploitation of a latent mechanism in graph contrastive learning: Representation scattering. Advances in Neural Information Processing Systems, 37:115351–115376, 2024

    2024

  3. [11]

    LightGCN: Simplifying and powering graph convolution network for recommendation

    Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. LightGCN: Simplifying and powering graph convolution network for recommendation. InProceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages ...

    2020

  4. [12]

    MixGCF: An improved training method for graph neural network-based recommender systems

    Tinglin Huang, Yuxiao Dong, Ming Ding, Zhen Yang, Wenzheng Feng, Xinyu Wang, and Jie Tang. MixGCF: An improved training method for graph neural network-based recommender systems. InProceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 665–674, 2021

    2021

  5. [13]

    Towards graph contrastive learning: A survey and beyond.ArXiv preprint arXiv:2405.11868, 2024

    Wei Ju, Yifan Wang, Yifang Qin, Zhengyang Mao, Zhiping Xiao, Junyu Luo, Junwei Yang, Yiyang Gu, Dongjie Wang, Qingqing Long, et al. Towards graph contrastive learning: A survey and beyond.ArXiv preprint arXiv:2405.11868, 2024

    2024 arXiv

  6. [14]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization.ArXiv preprint arXiv:1412.6980, 2014. 11

    2014 arXiv

  7. [15]

    Bootstrapping user and item representations for one-class collaborative filtering

    Dongha Lee, SeongKu Kang, Hyunjun Ju, Chanyoung Park, and Hwanjo Yu. Bootstrapping user and item representations for one-class collaborative filtering. InProceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 317–...

    2021

  8. [16]

    Hierarchical bipartite graph neural networks: Towards large-scale E-commerce applications

    Zhao Li, Xin Shen, Yuhang Jiao, Xuming Pan, Pengcheng Zou, Xianling Meng, Chengwei Yao, and Jiajun Bu. Hierarchical bipartite graph neural networks: Towards large-scale E-commerce applications. In2020 IEEE 36th International Conference on Data Engineering (ICDE), pages 1677–16...

    2020

  9. [17]

    Variational autoen- coders for collaborative filtering

    Dawen Liang, Rahul G Krishnan, Matthew D Hoffman, and Tony Jebara. Variational autoen- coders for collaborative filtering. InProceedings of the 2018 World Wide Web Conference, pages 689–698, 2018

    2018

  10. [18]

    Improving graph collaborative filtering with neighborhood-enriched contrastive learning

    Zihan Lin, Changxin Tian, Yupeng Hou, and Wayne Xin Zhao. Improving graph collaborative filtering with neighborhood-enriched contrastive learning. InProceedings of the ACM Web Conference 2022, pages 2320–2329, 2022

    2022

  11. [19]

    Representation learning with contrastive predictive coding.ArXiv preprint arXiv:1807.03748, 2018

    Aaron van den Oord, Yazhe Li, and Oriol Vinyals. Representation learning with contrastive predictive coding.ArXiv preprint arXiv:1807.03748, 2018

    2018 arXiv

  12. [20]

    BPR: Bayesian personalized ranking from implicit feedback.ArXiv preprint arXiv:1205.2618, 2012

    Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme. BPR: Bayesian personalized ranking from implicit feedback.ArXiv preprint arXiv:1205.2618, 2012

    2012 arXiv

  13. [21]

    Visualizing data using t-SNE.Journal of Machine Learning Research, 9(11), 2008

    Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-SNE.Journal of Machine Learning Research, 9(11), 2008

    2008

  14. [22]

    Deep graph infomax.ArXiv preprint arXiv:1809.10341, 2018

    Petar Veliˇckovi´c, William Fedus, William L Hamilton, Pietro Liò, Yoshua Bengio, and R Devon Hjelm. Deep graph infomax.ArXiv preprint arXiv:1809.10341, 2018

    2018 arXiv

  15. [23]

    Neural graph collabo- rative filtering

    Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. Neural graph collabo- rative filtering. InProceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 165–174, 2019

    2019

  16. [24]

    Self-supervised heterogeneous graph neural network with co-contrastive learning

    Xiao Wang, Nian Liu, Hui Han, and Chuan Shi. Self-supervised heterogeneous graph neural network with co-contrastive learning. InProceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 1726–1736, 2021

    2021

  17. [25]

    Self-supervised graph learning for recommendation

    Jiancan Wu, Xiang Wang, Fuli Feng, Xiangnan He, Liang Chen, Jianxun Lian, and Xing Xie. Self-supervised graph learning for recommendation. InProceedings of the 44th International ACM SIGIR Conference on Research and Development Information Retrieval, pages 726–735, 2021

    2021

  18. [26]

    Representing long-range context for graph neural networks with global attention

    Zhanghao Wu, Paras Jain, Matthew Wright, Azalia Mirhoseini, Joseph E Gonzalez, and Ion Stoica. Representing long-range context for graph neural networks with global attention. Advances in Neural Information Processing Systems, 34:13266–13279, 2021

    2021

  19. [27]

    Hyper- graph contrastive collaborative filtering

    Lianghao Xia, Chao Huang, Yong Xu, Jiashu Zhao, Dawei Yin, and Jimmy Huang. Hyper- graph contrastive collaborative filtering. InProceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 70–79, 2022

    2022

  20. [28]

    Simple and asymmetric graph contrastive learning without augmentations.Advances in Neural Information Processing Systems, 36:16129–16152, 2023

    Teng Xiao, Huaisheng Zhu, Zhengyu Chen, and Suhang Wang. Simple and asymmetric graph contrastive learning without augmentations.Advances in Neural Information Processing Systems, 36:16129–16152, 2023

    2023

  21. [29]

    InfoGCL: Information-aware graph contrastive learning.Advances in Neural Information Processing Systems, 34:30414–30425, 2021

    Dongkuan Xu, Wei Cheng, Dongsheng Luo, Haifeng Chen, and Xiang Zhang. InfoGCL: Information-aware graph contrastive learning.Advances in Neural Information Processing Systems, 34:30414–30425, 2021

    2021

  22. [30]

    Self-supervised graph- level representation learning with local and global structure

    Minghao Xu, Hang Wang, Bingbing Ni, Hongyu Guo, and Jian Tang. Self-supervised graph- level representation learning with local and global structure. InInternational Conference on Machine Learning, pages 11548–11558. PMLR, 2021. 12

    2021

  23. [31]

    Predicting individual irregular mobility via web search-driven bipartite graph neural networks.IEEE Transactions on Knowledge and Data Engineering, 2024

    Jiawei Xue, Takahiro Yabe, Kota Tsubouchi, Jianzhu Ma, and Satish V Ukkusuri. Predicting individual irregular mobility via web search-driven bipartite graph neural networks.IEEE Transactions on Knowledge and Data Engineering, 2024

    2024

  24. [32]

    Hierarchical graph contrastive learning

    Hao Yan, Senzhang Wang, Jun Yin, Chaozhuo Li, Junxing Zhu, and Jianxin Wang. Hierarchical graph contrastive learning. InJoint European Conference on Machine Learning and Knowledge Discovery in Databases, pages 700–715. Springer, 2023

    2023

  25. [33]

    An empirical study towards prompt-tuning for graph contrastive pre-training in recommendations.Advances in Neural Information Processing Systems, 36:62853–62868, 2023

    Haoran Yang, Xiangyu Zhao, Yicong Li, Hongxu Chen, and Guandong Xu. An empirical study towards prompt-tuning for graph contrastive pre-training in recommendations.Advances in Neural Information Processing Systems, 36:62853–62868, 2023

    2023

  26. [34]

    Self-supervised learning for large-scale item recommendations

    Tiansheng Yao, Xinyang Yi, Derek Zhiyuan Cheng, Felix Yu, Ting Chen, Aditya Menon, Lichan Hong, Ed H Chi, Steve Tjoa, Jieqi Kang, et al. Self-supervised learning for large-scale item recommendations. InProceedings of the 30th ACM International Conference on Information & Knowl...

    2021

  27. [35]

    Graph convolutional neural networks for web-scale recommender systems

    Rex Ying, Ruining He, Kaifeng Chen, Pong Eksombatchai, William L Hamilton, and Jure Leskovec. Graph convolutional neural networks for web-scale recommender systems. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 974–983, 2018

    2018

  28. [36]

    Hierarchical graph representation learning with differentiable pooling.Advances in Neural Information Processing Systems, 31, 2018

    Zhitao Ying, Jiaxuan You, Christopher Morris, Xiang Ren, Will Hamilton, and Jure Leskovec. Hierarchical graph representation learning with differentiable pooling.Advances in Neural Information Processing Systems, 31, 2018

    2018

  29. [37]

    Graph contrastive learning with augmentations.Advances in Neural Information Processing Systems, 33:5812–5823, 2020

    Yuning You, Tianlong Chen, Yongduo Sui, Ting Chen, Zhangyang Wang, and Yang Shen. Graph contrastive learning with augmentations.Advances in Neural Information Processing Systems, 33:5812–5823, 2020

    2020

  30. [38]

    XSimGCL: Towards extremely simple graph contrastive learning for recommendation.IEEE Transactions on Knowledge and Data Engineering, 36(2):913–926, 2023

    Junliang Yu, Xin Xia, Tong Chen, Lizhen Cui, Nguyen Quoc Viet Hung, and Hongzhi Yin. XSimGCL: Towards extremely simple graph contrastive learning for recommendation.IEEE Transactions on Knowledge and Data Engineering, 36(2):913–926, 2023

    2023

  31. [39]

    Are graph augmentations necessary? Simple graph contrastive learning for recommendation

    Junliang Yu, Hongzhi Yin, Xin Xia, Tong Chen, Lizhen Cui, and Quoc Viet Hung Nguyen. Are graph augmentations necessary? Simple graph contrastive learning for recommendation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Informatio...

    2022

  32. [40]

    Self-supervised learning for recommender systems: A survey.IEEE Transactions on Knowledge and Data Engineering, 36(1):335–355, 2023

    Junliang Yu, Hongzhi Yin, Xin Xia, Tong Chen, Jundong Li, and Zi Huang. Self-supervised learning for recommender systems: A survey.IEEE Transactions on Knowledge and Data Engineering, 36(1):335–355, 2023

    2023

  33. [41]

    ContextGNN: Beyond two-tower recommendation systems

    Yiwen Yuan, Zecheng Zhang, Xinwei He, Akihiro Nitta, Weihua Hu, Dong Wang, Manan Shah, Shenyang Huang, Blaž Stojanovi ˇc, Alan Krumholz, Jan Eric Lenssen, Jure Leskovec, and Matthias Fey. ContextGNN: Beyond two-tower recommendation systems. InThirteenth International Conferenc...

    2025

  34. [42]

    Long-tail augmented graph con- trastive learning for recommendation

    Qian Zhao, Zhengwei Wu, Zhiqiang Zhang, and Jun Zhou. Long-tail augmented graph con- trastive learning for recommendation. InJoint European Conference on Machine Learning and Knowledge Discovery in Databases, pages 387–403. Springer, 2023

    2023

  35. [43]

    Deep graph contrastive representation learning.ArXiv preprint arXiv:2006.04131, 2020

    Yanqiao Zhu, Yichen Xu, Feng Yu, Qiang Liu, Shu Wu, and Liang Wang. Deep graph contrastive representation learning.ArXiv preprint arXiv:2006.04131, 2020. 13

    2006 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.