{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/distilling-self-knowledge-from-contrastive","title":"Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages","arxiv_id":"2106.08541","date":"2021-06-16","proceeding":null,"authors":["Yi Luo","Aiguo Chen","Ke Yan","Ling Tian"],"abstract":"Nowadays, Graph Neural Networks (GNNs) following the Message Passing paradigm become the dominant way to learn on graphic data. Models in this paradigm have to spend extra space to look up adjacent nodes with adjacency matrices and extra time to aggregate multiple messages from adjacent nodes. To address this issue, we develop a method called LinkDist that distils self-knowledge from connected node pairs into a Multi-Layer Perceptron (MLP) without the need to aggregate messages. Experiment with 8 real-world datasets shows the MLP derived from LinkDist can predict the label of a node without knowing its adjacencies but achieve comparable accuracy against GNNs in the contexts of semi- and full-supervised node classification. Moreover, LinkDist benefits from its Non-Message Passing paradigm that we can also distil self-knowledge from arbitrarily sampled node pairs in a contrastive way to further boost the performance of LinkDist.","url_abs":"https://arxiv.org/abs/2106.08541v1","url_pdf":"https://arxiv.org/pdf/2106.08541v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"distilling-self-knowledge-from-contrastive","repo_url":"https://github.com/cf020031308/LinkDist","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"distilling-self-knowledge-from-contrastive","repo_url":"https://github.com/cf020031308/LinkDist/blob/master/ogbn.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"node-classification","task_name":"Node Classification"},{"task_slug":"node-property-prediction","task_name":"Node Property Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/node-classification-on-amazon-computers-1","task":"Node Classification","dataset":"Amazon Computers","model":"LinkDist","rank_in_archive_order":9,"of":12,"metrics":{"Accuracy":"89.49%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-computers-1","task":"Node Classification","dataset":"Amazon Computers","model":"LinkDistMLP","rank_in_archive_order":10,"of":12,"metrics":{"Accuracy":"89.44%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-computers-1","task":"Node Classification","dataset":"Amazon Computers","model":"CoLinkDist","rank_in_archive_order":11,"of":12,"metrics":{"Accuracy":"89.42%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-computers-1","task":"Node Classification","dataset":"Amazon Computers","model":"CoLinkDistMLP","rank_in_archive_order":12,"of":12,"metrics":{"Accuracy":"88.85%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-photo-1","task":"Node Classification","dataset":"Amazon Photo","model":"CoLinkDist","rank_in_archive_order":8,"of":11,"metrics":{"Accuracy":"94.36%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-photo-1","task":"Node Classification","dataset":"Amazon Photo","model":"CoLinkDistMLP","rank_in_archive_order":9,"of":11,"metrics":{"Accuracy":"94.12%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-photo-1","task":"Node Classification","dataset":"Amazon Photo","model":"LinkDistMLP","rank_in_archive_order":10,"of":11,"metrics":{"Accuracy":"93.83%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-amazon-photo-1","task":"Node Classification","dataset":"Amazon Photo","model":"LinkDist","rank_in_archive_order":11,"of":11,"metrics":{"Accuracy":"93.75%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer-with-public","task":"Node Classification","dataset":"CiteSeer with Public Split: fixed 20 nodes per class","model":"CoLinkDistMLP","rank_in_archive_order":29,"of":40,"metrics":{"Accuracy":"70.96%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer-with-public","task":"Node Classification","dataset":"CiteSeer with Public Split: fixed 20 nodes per class","model":"CoLinkDist","rank_in_archive_order":30,"of":40,"metrics":{"Accuracy":"70.79%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer-with-public","task":"Node Classification","dataset":"CiteSeer with Public Split: fixed 20 nodes per class","model":"LinkDist","rank_in_archive_order":31,"of":40,"metrics":{"Accuracy":"70.27%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer-with-public","task":"Node Classification","dataset":"CiteSeer with Public Split: fixed 20 nodes per class","model":"LinkDistMLP","rank_in_archive_order":32,"of":40,"metrics":{"Accuracy":"70.26%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"CoLinkDist","rank_in_archive_order":17,"of":71,"metrics":{"Accuracy":"75.79%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"CoLinkDistMLP","rank_in_archive_order":18,"of":71,"metrics":{"Accuracy":"75.77%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"LinkDistMLP","rank_in_archive_order":22,"of":71,"metrics":{"Accuracy":"75.25%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"LinkDist","rank_in_archive_order":24,"of":71,"metrics":{"Accuracy":"74.72%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-cs","task":"Node Classification","dataset":"Coauthor CS","model":"CoLinkDist","rank_in_archive_order":6,"of":24,"metrics":{"Accuracy":"95.80%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-cs","task":"Node Classification","dataset":"Coauthor CS","model":"CoLinkDistMLP","rank_in_archive_order":7,"of":24,"metrics":{"Accuracy":"95.74%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-cs","task":"Node Classification","dataset":"Coauthor CS","model":"LinkDistMLP","rank_in_archive_order":9,"of":24,"metrics":{"Accuracy":"95.68%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-cs","task":"Node Classification","dataset":"Coauthor CS","model":"LinkDist","rank_in_archive_order":11,"of":24,"metrics":{"Accuracy":"95.66%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-physics","task":"Node Classification","dataset":"Coauthor Physics","model":"CoLinkDist","rank_in_archive_order":5,"of":14,"metrics":{"Accuracy":"97.05%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-physics","task":"Node Classification","dataset":"Coauthor Physics","model":"LinkDistMLP","rank_in_archive_order":8,"of":14,"metrics":{"Accuracy":"96.91%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-physics","task":"Node Classification","dataset":"Coauthor Physics","model":"CoLinkDistMLP","rank_in_archive_order":10,"of":14,"metrics":{"Accuracy":"96.87%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-coauthor-physics","task":"Node Classification","dataset":"Coauthor Physics","model":"LinkDist","rank_in_archive_order":11,"of":14,"metrics":{"Accuracy":"96.87%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"LinkDist","rank_in_archive_order":10,"of":73,"metrics":{"Accuracy":"88.24%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"CoLinkDist","rank_in_archive_order":12,"of":73,"metrics":{"Accuracy":"87.89%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"LinkDistMLP","rank_in_archive_order":14,"of":73,"metrics":{"Accuracy":"87.58%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"CoLinkDistMLP","rank_in_archive_order":15,"of":73,"metrics":{"Accuracy":"87.54%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full","task":"Node Classification","dataset":"Cora Full","model":"CoLinkDist","rank_in_archive_order":2,"of":5,"metrics":{"Accuracy":"70.32%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full","task":"Node Classification","dataset":"Cora Full","model":"LinkDist","rank_in_archive_order":3,"of":5,"metrics":{"Accuracy":"69.87%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full","task":"Node Classification","dataset":"Cora Full","model":"CoLinkDistMLP","rank_in_archive_order":4,"of":5,"metrics":{"Accuracy":"69.83%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full","task":"Node Classification","dataset":"Cora Full","model":"LinkDistMLP","rank_in_archive_order":5,"of":5,"metrics":{"Accuracy":"69.53%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full-with-public","task":"Node Classification","dataset":"Cora Full with Public Split","model":"CoLinkDist","rank_in_archive_order":1,"of":4,"metrics":{"Accuracy":"57.05%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full-with-public","task":"Node Classification","dataset":"Cora Full with Public Split","model":"LinkDist","rank_in_archive_order":2,"of":4,"metrics":{"Accuracy":"55.87%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full-with-public","task":"Node Classification","dataset":"Cora Full with Public Split","model":"CoLinkDistMLP","rank_in_archive_order":3,"of":4,"metrics":{"Accuracy":"53.43%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-full-with-public","task":"Node Classification","dataset":"Cora Full with Public Split","model":"LinkDistMLP","rank_in_archive_order":4,"of":4,"metrics":{"Accuracy":"51.78%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-with-public-split","task":"Node Classification","dataset":"Cora with Public Split: fixed 20 nodes per class","model":"CoLinkDist","rank_in_archive_order":26,"of":36,"metrics":{"Accuracy":"81.39%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-with-public-split","task":"Node Classification","dataset":"Cora with Public Split: fixed 20 nodes per class","model":"CoLinkDistMLP","rank_in_archive_order":27,"of":36,"metrics":{"Accuracy":"81.19%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-with-public-split","task":"Node Classification","dataset":"Cora with Public Split: fixed 20 nodes per class","model":"LinkDist","rank_in_archive_order":28,"of":36,"metrics":{"Accuracy":"81.05%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-with-public-split","task":"Node Classification","dataset":"Cora with Public Split: fixed 20 nodes per class","model":"LinkDistMLP","rank_in_archive_order":29,"of":36,"metrics":{"Accuracy":"80.79%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-with-public","task":"Node Classification","dataset":"PubMed with Public Split: fixed 20 nodes per class","model":"CoLinkDist","rank_in_archive_order":33,"of":37,"metrics":{"Accuracy":"75.64%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-with-public","task":"Node Classification","dataset":"PubMed with Public Split: fixed 20 nodes per class","model":"CoLinkDistMLP","rank_in_archive_order":34,"of":37,"metrics":{"Accuracy":"75.41%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-with-public","task":"Node Classification","dataset":"PubMed with Public Split: fixed 20 nodes per class","model":"LinkDist","rank_in_archive_order":35,"of":37,"metrics":{"Accuracy":"74.06%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-with-public","task":"Node Classification","dataset":"PubMed with Public Split: fixed 20 nodes per class","model":"LinkDistMLP","rank_in_archive_order":36,"of":37,"metrics":{"Accuracy":"72.41%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"CoLinkDist","rank_in_archive_order":9,"of":70,"metrics":{"Accuracy":"89.58%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"CoLinkDistMLP","rank_in_archive_order":10,"of":70,"metrics":{"Accuracy":"89.53%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"LinkDist","rank_in_archive_order":14,"of":70,"metrics":{"Accuracy":"88.86%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"LinkDistMLP","rank_in_archive_order":15,"of":70,"metrics":{"Accuracy":"88.79%"},"uses_additional_data":false},{"leaderboard":"/sota/node-property-prediction-on-ogbn-arxiv","task":"Node Property Prediction","dataset":"ogbn-arxiv","model":"CoLinkDistMLP","rank_in_archive_order":83,"of":86,"metrics":{"Ext. data":"No","Number of params":"120912","Test Accuracy":"0.5638 ± 0.0016","Validation Accuracy":"0.5807 ± 0.0011"},"uses_additional_data":false},{"leaderboard":"/sota/node-property-prediction-on-ogbn-mag","task":"Node Property Prediction","dataset":"ogbn-mag","model":"CoLinkDistMLP","rank_in_archive_order":38,"of":39,"metrics":{"Ext. data":"No","Number of params":"278202","Test Accuracy":"0.2761 ± 0.0018","Validation Accuracy":"0.2646 ± 0.0013"},"uses_additional_data":false},{"leaderboard":"/sota/node-property-prediction-on-ogbn-products","task":"Node Property Prediction","dataset":"ogbn-products","model":"CoLinkDistMLP","rank_in_archive_order":62,"of":64,"metrics":{"Ext. data":"No","Number of params":"115806","Test Accuracy":"0.6259 ± 0.0010","Validation Accuracy":"0.7721 ± 0.0015"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.08541","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.08541"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cf020031308/LinkDist","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cf020031308/LinkDist/blob/master/ogbn.py","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"eaab7909ebfb427e","entry":"optimize","repo":"cf020031308/LinkDist","repo_kind":"official","path":"ogbn.py","file_url":"https://github.com/cf020031308/LinkDist/blob/HEAD/ogbn.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"eaab7909ebfb427e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}