pith. sign in

arxiv: 2507.05311 · v2 · pith:LEHGBQDVnew · submitted 2025-07-07 · 💻 cs.IR · cs.AI

PLACE: Prompt Learning for Attributed Community Search in Large Graphs

Shuheng Fang , Kangfei Zhao , Rener Zhang , Yu Rong , Jeffrey Xu Yu This is my paper

Pith reviewed 2026-05-25 07:45 UTC · model grok-4.3

classification 💻 cs.IR cs.AI
keywords prompt learningattributed community searchgraph neural networkscommunity detectionlarge-scale graphsquery-dependent refinement
0
0 comments X

The pith

Learnable prompt tokens inserted into graphs act as bridges to let GNNs identify attributed communities matching specific queries.

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

The paper sets out to adapt prompt-tuning ideas from language models to attributed community search on graphs. It builds a prompt-augmented graph by adding both structural tokens and learnable prompt tokens that depend on the query. These prompt tokens are positioned to strengthen node connections so that a graph neural network can more readily pick up the mix of structure and attributes that define the desired community. Training alternates between updating the prompt parameters and the GNN weights, while a divide-and-conquer step keeps the method workable on very large graphs. The authors test the resulting model on nine real graphs across three query types.

Core claim

The central claim is that integrating structural and learnable prompt tokens into the graph forms a prompt-augmented graph in which the learned prompt tokens serve as a bridge that strengthens connections between graph nodes for the query. This bridge enables the GNN to more effectively identify patterns of structural cohesiveness and attribute similarity related to the specific query. An alternating training paradigm optimizes both the prompt parameters and the GNN jointly, and a divide-and-conquer strategy supports handling million-scale graphs.

What carries the argument

The prompt-augmented graph, created by inserting learnable prompt tokens as a query-dependent refinement mechanism that bridges node connections.

If this is right

  • The GNN gains the ability to focus on query-specific cohesiveness and attribute patterns through the strengthened connections.
  • Joint alternating optimization lets the prompt tokens and the GNN adapt together to the community search task.
  • The divide-and-conquer strategy allows the framework to scale to graphs with millions of nodes.
  • The same prompt-insertion approach applies across multiple types of attributed community search queries.

Where Pith is reading between the lines

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

  • The same insertion of learnable tokens could be tried on other graph tasks that involve matching a query to local structure and attributes.
  • If the bridge effect holds, one could test whether the prompts remain useful when the underlying GNN architecture is swapped for a different one.
  • Extending the alternating training to graphs that change over time might let communities update as new queries arrive.

Load-bearing premise

Learnable prompt tokens inserted into the graph will serve as an effective bridge that strengthens connections between nodes for the query.

What would settle it

Running the same GNN on the nine real-world graphs and three query types but without the inserted prompt tokens, and finding equal or higher community detection accuracy, would show the bridge mechanism adds no value.

Figures

Figures reproduced from arXiv: 2507.05311 by Jeffrey Xu Yu, Kangfei Zhao, Rener Zhang, Shuheng Fang, Yu Rong.

Figure 1
Figure 1. Figure 1: Query prompt inspired by the prompt of LLMs. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architecture of PLACE: Given a query, prompt tokens are inserted into the original graph for context enrichment, forming a prompt-augmented graph. Then, the prompt-augmented graph is fed into a GNN-based encoder-encoder neural network for predicting the community membership of the query. Given a set of training queries 𝑄 and ground-truth 𝐿, ACS is for￾mulated as a query-specific binary classification task … view at source ↗
Figure 3
Figure 3. Figure 3: Construction query-prompt graph given a query [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Alternative Training of Prompt Encoder and GNN [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of Training and Test Time (s) [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: F1 Score under Different GNN Layers AFC AFN EQA 0.5 0.6 0.7 0.8 0.9 w/o fine-tune fine-tune prompt fine-tune prompt & GNN AFC AFN EQA 0.5 0.6 0.7 0.8 F1 (a) Train: Cor. Test: Tex. AFC AFN EQA 0.5 0.6 0.7 0.8 0.9 (b) Train: Wis. Test: Tex. AFC AFN EQA 0.5 0.6 0.7 0.8 (c) Train: Cor. Test: Was [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 6
Figure 6. Figure 6: F1 Score of PLACE under Different Number of Virtual Node and Ratio of Positive/Negative Samples sample shards in one epoch instead of training the entire graph, which reflects our scalability. 7.3.2 Test time [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 9
Figure 9. Figure 9: Visualization of embeddings for nodes in prompt [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
read the original abstract

In this paper, we propose PLACE (Prompt Learning for Attributed Community Search), an innovative graph prompt learning framework for ACS. Enlightened by prompt-tuning in Natural Language Processing (NLP), where learnable prompt tokens are inserted to contextualize NLP queries, PLACE integrates structural and learnable prompt tokens into the graph as a query-dependent refinement mechanism, forming a prompt-augmented graph. Within this prompt-augmented graph structure, the learned prompt tokens serve as a bridge that strengthens connections between graph nodes for the query, enabling the GNN to more effectively identify patterns of structural cohesiveness and attribute similarity related to the specific query. We employ an alternating training paradigm to optimize both the prompt parameters and the GNN jointly. Moreover, we design a divide-and-conquer strategy to enhance scalability, supporting the model to handle million-scale graphs. Extensive experiments on 9 real-world graphs demonstrate the effectiveness of PLACE for three types of ACS queries, where PLACE achieves higher F1 scores by 22% compared to the state-of-the-arts on average.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The manuscript proposes PLACE, a graph prompt learning framework for attributed community search (ACS). It inserts structural and learnable prompt tokens into the input graph to create a prompt-augmented graph in which the tokens act as query-dependent bridges that strengthen connections, enabling a GNN to better capture structural cohesiveness and attribute similarity. The method optimizes prompt embeddings and GNN parameters via alternating training and employs a divide-and-conquer strategy for scalability to million-scale graphs. Experiments on nine real-world graphs report an average 22% F1 improvement over state-of-the-art baselines across three ACS query types.

Significance. If the reported F1 gains hold under standard experimental controls, the work would contribute a concrete adaptation of NLP-style prompt tuning to attributed community search, with the prompt-augmented graph construction and alternating training as the core technical ideas. The scalability component is presented separately from the performance numbers and addresses a practical requirement for large graphs. The manuscript supplies an explicit mechanistic hypothesis (prompt tokens as bridges) that is directly tied to the empirical claim.

minor comments (2)
  1. [Abstract] Abstract: the phrase 'state-of-the-arts' should read 'state-of-the-art methods' for standard English usage.
  2. [Abstract] Abstract: the description of the prompt-augmented graph mechanism is concise but would benefit from a parenthetical reference to the relevant figure or section that illustrates the token insertion and bridging effect.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, significance assessment, and recommendation for minor revision. We appreciate the recognition of the prompt-augmented graph construction, alternating training, and scalability strategy as core contributions. No major comments appear in the provided report, so we have no specific points requiring rebuttal or revision at this stage.

Circularity Check

0 steps flagged

No significant circularity; empirical performance claims only

full rationale

The paper introduces PLACE as a prompt-augmented graph framework for attributed community search, relying on alternating training of prompts and GNN plus a divide-and-conquer scalability strategy. All load-bearing claims are empirical (average 22% F1 improvement across 9 graphs and three query types) and are validated against external baselines rather than derived from internal equations. No self-definitional reductions, fitted inputs renamed as predictions, or load-bearing self-citations appear in the provided text. The prompt-token bridge mechanism is an explicit modeling choice presented as validated by experiment, not a derivation that collapses to its own inputs. The work is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 1 invented entities

The claim rests on the effectiveness of learnable prompt tokens as bridges in graphs and the alternating optimization; free parameters include the prompt token embeddings and GNN weights fitted during training; axioms include standard assumptions that GNNs can exploit augmented graph structures for community detection.

free parameters (2)
  • learnable prompt token embeddings
    Parameters optimized to contextualize queries in the prompt-augmented graph.
  • GNN model parameters
    Weights trained jointly via alternating paradigm on the augmented graphs.
axioms (1)
  • domain assumption GNNs can learn patterns of structural cohesiveness and attribute similarity from prompt-augmented graphs
    Invoked to justify why the prompt tokens enable better identification of communities.
invented entities (1)
  • prompt tokens no independent evidence
    purpose: To act as query-dependent bridges strengthening node connections in the graph
    New learnable entities introduced to form the prompt-augmented graph; no independent evidence provided outside the framework itself.

pith-pipeline@v0.9.0 · 5718 in / 1366 out tokens · 33305 ms · 2026-05-25T07:45:09.968515+00:00 · methodology

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

64 extracted references · 64 canonical work pages · 1 internal anchor

  1. [1]

    [n.d.]. Pytorch. https://github.com/pytorch/pytorch

    work page

  2. [2]

    Pytorch Geometric

    [n.d.]. Pytorch Geometric. https://github.com/rusty1s/pytorch_geometric

    work page

  3. [3]

    Esra Akbas and Peixiang Zhao. 2017. Truss-based community search: a truss- equivalence based indexing approach. Proceedings of the VLDB Endowment 10, 11 (2017), 1298–1309

    work page 2017

  4. [4]

    Jiazun Chen, Jun Gao, and Bin Cui. 2023. ICS-GNN +: lightweight interactive community search via graph neural network. VLDB J. 32, 2 (2023), 447–467. https://doi.org/10.1007/S00778-022-00754-0

    work page doi:10.1007/s00778-022-00754-0 2023

  5. [5]

    Jiazun Chen, Yikuan Xia, and Jun Gao. 2023. CommunityAF: An Example-Based Community Search Method via Autoregressive Flow. Proceedings of the VLDB Endowment 16, 10 (2023), 2565–2577

    work page 2023

  6. [6]

    Wei-Lin Chiang, Xuanqing Liu, Si Si, Yang Li, Samy Bengio, and Cho-Jui Hsieh

    work page

  7. [7]

    In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining

    Cluster-gcn: An efficient algorithm for training deep and large graph convolutional networks. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining . 257–266

    work page

  8. [8]

    Wanyun Cui, Yanghua Xiao, Haixun Wang, Yiqi Lu, and Wei Wang. 2013. Online search of overlapping communities. In Proceedings of the 2013 ACM SIGMOD international conference on Management of data . 277–288

    work page 2013

  9. [9]

    Wanyun Cui, Yanghua Xiao, Haixun Wang, and Wei Wang. 2014. Local search of communities in large graphs. In Proceedings of the 2014 ACM SIGMOD interna- tional conference on Management of data . 991–1002

    work page 2014

  10. [10]

    Shuheng Fang, Kangfei Zhao, Guanghua Li, and Jeffrey Xu Yu. 2023. Community search: a meta-learning approach. In 2023 IEEE 39th International Conference on Data Engineering (ICDE). IEEE, 2358–2371

    work page 2023

  11. [11]

    Shuheng Fang, Kangfei Zhao, Yu Rong, Zhixun Li, and Jeffrey Xu Yu. 2024. Inductive Attributed Community Search: To Learn Communities Across Graphs. Proceedings of the VLDB Endowment 17, 10 (2024), 2576–2589

    work page 2024

  12. [12]

    Yixiang Fang, CK Cheng, Siqiang Luo, and Jiafeng Hu. 2016. Effective community search for large attributed graphs. Proceedings of the VLDB Endowment (2016)

    work page 2016

  13. [13]

    Yixiang Fang, Xin Huang, Lu Qin, Ying Zhang, Wenjie Zhang, Reynold Cheng, and Xuemin Lin. 2020. A survey of community search over big graphs.The VLDB Journal 29 (2020), 353–392

    work page 2020

  14. [14]

    Jun Gao, Jiazun Chen, Zhao Li, and Ji Zhang. 2021. ICS-GNN: lightweight interactive community search via graph neural network. Proceedings of the VLDB Endowment 14, 6 (2021), 1006–1018

    work page 2021

  15. [15]

    Chenghua Gong, Xiang Li, Jianxiang Yu, Cheng Yao, Jiaqi Tan, Chengcheng Yu, and Dawei Yin. 2023. Prompt tuning for multi-view graph contrastive learning. arXiv preprint arXiv:2310.10362 (2023)

    work page arXiv 2023

  16. [16]

    Fangda Guo, Ye Yuan, Guoren Wang, Xiangguo Zhao, and Hao Sun. 2021. Multi- attributed community search in road-social networks. In 2021 IEEE 37th Interna- tional Conference on Data Engineering (ICDE) . IEEE, 109–120

    work page 2021

  17. [17]

    Will Hamilton, Zhitao Ying, and Jure Leskovec. 2017. Inductive representation learning on large graphs. Advances in neural information processing systems 30 (2017)

    work page 2017

  18. [18]

    Xin Huang, Hong Cheng, Lu Qin, Wentao Tian, and Jeffrey Xu Yu. 2014. Querying k-truss community in large and dynamic graphs. In Proc. SIGMOD. ACM, 1311– 1322

    work page 2014

  19. [19]

    Xin Huang, Laks VS Lakshmanan, Jeffrey Xu Yu, and Hong Cheng. 2015. Approx- imate closest community search in networks. arXiv preprint arXiv:1505.05956 (2015)

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  20. [20]

    Xin Huang and Laks V. S. Lakshmanan. 2017. Attribute-Driven Community Search. Proc. VLDB Endow. 10, 9 (2017), 949–960

    work page 2017

  21. [21]

    Belongie, Bharath Hariharan, and Ser-Nam Lim

    Menglin Jia, Luming Tang, Bor-Chun Chen, Claire Cardie, Serge J. Belongie, Bharath Hariharan, and Ser-Nam Lim. 2022. Visual Prompt Tuning. In Computer Vision - ECCV 2022 - 17th European Conference, Tel A viv, Israel, October 23-27, 2022, Proceedings, Part XXXIII (Lecture Notes in Computer Science) , Vol. 13693. Springer, 709–727. https://doi.org/10.1007/9...

    work page doi:10.1007/978-3-031-19827-4_41 2022

  22. [22]

    Yuli Jiang, Yu Rong, Hong Cheng, Xin Huang, Kangfei Zhao, and Junzhou Huang

    work page

  23. [23]

    Proceedings of the VLDB Endow- ment 15, 6 (2022), 1243–1255

    Query driven-graph neural networks for community search: from non- attributed, attributed, to interactive attributed. Proceedings of the VLDB Endow- ment 15, 6 (2022), 1243–1255

    work page 2022

  24. [24]

    George Karypis and Vipin Kumar. 1997. METIS: A software package for parti- tioning unstructured graphs, partitioning meshes, and computing fill-reducing orderings of sparse matrices. (1997)

    work page 1997

  25. [25]

    Juyong Lee and Jooyoung Lee. 2013. Hidden information revealed by optimal community structure from a protein-complex bipartite network improves protein function prediction. PloS one 8, 4 (2013), e60372

    work page 2013

  26. [26]

    Jure Leskovec and Julian Mcauley. 2012. Learning to discover social circles in ego networks. Advances in neural information processing systems 25 (2012)

    work page 2012

  27. [27]

    Brian Lester, Rami Al-Rfou, and Noah Constant. 2021. The Power of Scale for Parameter-Efficient Prompt Tuning. In Proceedings of the 2021 Conference on Em- pirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 7-11 November, 2021 . Association for Computational Linguistics, 3045–3059. https://doi.org/1...

    work page doi:10.18653/v1/2021.emnlp-main.243 2021

  28. [28]

    Ling Li, Siqiang Luo, Yuhai Zhao, Caihua Shan, Zhengkui Wang, and Lu Qin

    work page

  29. [29]

    IEEE 39th ICDE (2023)

    COCLEP: Contrastive Learning-based Semi-Supervised Community Search. IEEE 39th ICDE (2023)

    work page 2023

  30. [30]

    Rong-Hua Li, Lu Qin, Jeffrey Xu Yu, and Rui Mao. 2015. Influential community search in large networks. Proceedings of the VLDB Endowment 8, 5 (2015), 509– 520

    work page 2015

  31. [31]

    Qing Liu, Yifan Zhu, Minjun Zhao, Xin Huang, Jianliang Xu, and Yunjun Gao. 2020. VAC: vertex-centric attributed community search. In2020 IEEE 36th International Conference on Data Engineering (ICDE) . IEEE, 937–948

    work page 2020

  32. [32]

    Xiao Liu, Kaixuan Ji, Yicheng Fu, Weng Tam, Zhengxiao Du, Zhilin Yang, and Jie Tang. 2022. P-Tuning: Prompt Tuning Can Be Comparable to Fine-tuning Across Scales and Tasks. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), ACL 2022, Dublin, Ireland, May 22-27, 2022, Smaranda Muresan, Presl...

    work page doi:10.18653/v1/ 2022

  33. [33]

    Zemin Liu, Xingtong Yu, Yuan Fang, and Xinming Zhang. 2023. Graphprompt: Unifying pre-training and downstream tasks for graph neural networks. In Pro- ceedings of the ACM Web Conference 2023 . 417–428

    work page 2023

  34. [34]

    Jiehuan Luo, Xin Cao, Xike Xie, Qiang Qu, Zhiqiang Xu, and Christian S Jensen

    work page

  35. [35]

    In 2020 IEEE 36th International Conference on Data Engineering (ICDE)

    Efficient attribute-constrained co-located community search. In 2020 IEEE 36th International Conference on Data Engineering (ICDE) . IEEE, 1201–1212

    work page 2020

  36. [36]

    Sungho Park and Hyeran Byun. 2024. Fair-VPT: Fair Visual Prompt Tuning for Image Classification. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, W A, USA, June 16-22, 2024. IEEE, 12268–12278. https://doi.org/10.1109/CVPR52733.2024.01166

    work page doi:10.1109/cvpr52733.2024.01166 2024

  37. [37]

    Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, and Bo Yang

    work page

  38. [38]

    arXiv preprint arXiv:2002.05287 (2020)

    Geom-gcn: Geometric graph convolutional networks. arXiv preprint arXiv:2002.05287 (2020)

    work page arXiv 2002

  39. [39]

    Dhiman Sarma, Wahidul Alam, Ishita Saha, Mohammad Nazmul Alam, Moham- mad Jahangir Alam, and Sohrab Hossain. 2020. Bank fraud detection using community detection algorithm. In 2020 second international conference on inven- tive research in computing applications (ICIRCA) . IEEE, 642–646

    work page 2020

  40. [40]

    Michael Schlichtkrull, Thomas N Kipf, Peter Bloem, Rianne Van Den Berg, Ivan Titov, and Max Welling. 2018. Modeling relational data with graph convolu- tional networks. In The semantic web: 15th international conference, ESWC 2018, Heraklion, Crete, Greece, June 3–7, 2018, proceedings 15 . Springer, 593–607

    work page 2018

  41. [41]

    Reza Shirkavand and Heng Huang. 2023. Deep prompt tuning for graph trans- formers. arXiv preprint arXiv:2309.10131 (2023)

    work page arXiv 2023

  42. [42]

    Yixin Song, Lihua Zhou, Peizhong Yang, Jialong Wang, and Lizhen Wang

    work page

  43. [43]

    IEEE Trans

    CS-DAHIN: Community Search Over Dynamic Attribute Heterogeneous Network. IEEE Trans. Knowl. Data Eng. 36, 11 (2024), 5874–5888. https: //doi.org/10.1109/TKDE.2024.3402258

    work page doi:10.1109/tkde.2024.3402258 2024

  44. [44]

    Mauro Sozio and Aristides Gionis. 2010. The community-search problem and how to plan a successful cocktail party. In Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining . 939–948

    work page 2010

  45. [45]

    Mingchen Sun, Kaixiong Zhou, Xin He, Ying Wang, and Xin Wang. 2022. Gppt: Graph pre-training and prompt tuning to generalize graph neural networks. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1717–1727

    work page 2022

  46. [46]

    Xiangguo Sun, Hong Cheng, Jia Li, Bo Liu, and Jihong Guan. 2023. All in one: Multi-task prompting for graph neural networks. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 2120–2131

    work page 2023

  47. [47]

    Xiangguo Sun, Jiawen Zhang, Xixi Wu, Hong Cheng, Yun Xiong, and Jia Li. 2023. Graph prompt learning: A comprehensive survey and beyond. arXiv preprint arXiv:2311.16534 (2023)

    work page arXiv 2023

  48. [48]

    Zhen Tan, Ruocheng Guo, Kaize Ding, and Huan Liu. 2023. Virtual node tuning for few-shot node classification. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 2177–2188

    work page 2023

  49. [49]

    Jiaxi Tang and Ke Wang. 2018. Personalized top-n sequential recommenda- tion via convolutional sequence embedding. In Proceedings of the eleventh ACM international conference on web search and data mining . 565–573

    work page 2018

  50. [50]

    Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. Journal of machine learning research 9, 11 (2008)

    work page 2008

  51. [51]

    Vasimuddin, Sanchit Misra, Guixiang Ma, Ramanarayan Mohanty, Evangelos Georganas, Alexander Heinecke, Dhiraj D

    Md. Vasimuddin, Sanchit Misra, Guixiang Ma, Ramanarayan Mohanty, Evangelos Georganas, Alexander Heinecke, Dhiraj D. Kalamkar, Nesreen K. Ahmed, and Sasikanth Avancha. 2021. DistGNN: scalable distributed training for large-scale graph neural networks. In International Conference for High Performance Comput- ing, Networking, Storage and Analysis, SC 2021, S...

    work page doi:10.1145/3458817.3480856 2021

  52. [52]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. In Advances in Neural Information Processing Systems 30: Annual Con- ference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA . 5998–6008. https://proceedings.n...

    work page 2017

  53. [53]

    Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, Yoshua Bengio, et al. 2017. Graph attention networks. stat 1050, 20 (2017), 10–48550

    work page 2017

  54. [54]

    Jianwei Wang, Kai Wang, Xuemin Lin, Wenjie Zhang, and Ying Zhang. 2024. Efficient Unsupervised Community Search with Pre-trained Graph Transformer. arXiv preprint arXiv:2403.18869 (2024)

    work page arXiv 2024

  55. [55]

    Jianwei Wang, Kai Wang, Xuemin Lin, Wenjie Zhang, and Ying Zhang. 2024. Neural Attributed Community Search at Billion Scale. Proceedings of the ACM on Management of Data 1, 4 (2024), 1–25

    work page 2024

  56. [56]

    Zonghan Wu, Shirui Pan, Fengwen Chen, Guodong Long, Chengqi Zhang, and Philip S. Yu. 2021. A Comprehensive Survey on Graph Neural Networks. IEEE Trans. Neural Networks Learn. Syst. 32, 1 (2021), 4–24. https://doi.org/10.1109/ TNNLS.2020.2978386

    work page arXiv 2021

  57. [57]

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. 2019. How Powerful are Graph Neural Networks?. In 7th International Conference on Learning Rep- resentations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019 . OpenReview.net. https://openreview.net/forum?id=ryGs6iA5Km

    work page 2019

  58. [58]

    Zhilin Yang, William Cohen, and Ruslan Salakhudinov. 2016. Revisiting semi- supervised learning with graph embeddings. In International conference on ma- chine learning. PMLR, 40–48

    work page 2016

  59. [59]

    Fajie Yuan, Alexandros Karatzoglou, Ioannis Arapakis, Joemon M Jose, and Xi- angnan He. 2019. A simple convolutional generative network for next item recommendation. In Proceedings of the twelfth ACM international conference on web search and data mining . 582–590

    work page 2019

  60. [60]

    Long Yuan, Lu Qin, Wenjie Zhang, Lijun Chang, and Jianye Yang. 2017. Index- based densest clique percolation community search in networks. IEEE Transac- tions on Knowledge and Data Engineering 30, 5 (2017), 922–935

    work page 2017

  61. [61]

    Prasanna

    Hanqing Zeng, Hongkuan Zhou, Ajitesh Srivastava, Rajgopal Kannan, and Viktor K. Prasanna. 2020. GraphSAINT: Graph Sampling Based Inductive Learning Method. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020 . OpenReview.net. https: //openreview.net/forum?id=BJe8pkHFwS

    work page 2020

  62. [62]

    Haihong Zhao, Aochuan Chen, Xiangguo Sun, Hong Cheng, and Jia Li. 2024. All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretraining. In Proc. KDD . ACM, 4443–4454. https://doi.org/10.1145/ 3637528.3671913

    work page arXiv 2024

  63. [63]

    Da Zheng, Chao Ma, Minjie Wang, Jinjing Zhou, Qidong Su, Xiang Song, Quan Gan, Zheng Zhang, and George Karypis. 2020. DistDGL: Distributed graph neural network training for billion-scale graphs. In 2020 IEEE/ACM 10th Workshop on Irregular Applications: Architectures and Algorithms (IA3) . IEEE, 36–44

    work page 2020

  64. [64]

    Yun Zhu, Jianhao Guo, and Siliang Tang. 2023. Sgl-pt: A strong graph learner with graph prompt tuning. arXiv preprint arXiv:2302.12449 (2023)

    work page arXiv 2023