Papers › Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without...

Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages

16 Jun 2021arXiv:2106.08541archive 2025-07-28

Yi Luo, Aiguo Chen, Ke Yan, Ling Tian

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.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 1 code sample harvested from 1 repository linked to this paper: 0 ran · 1 unverified.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

cf020031308/LinkDist officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
cf020031308/LinkDist mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

1 sample harvested: 0 ran · 1 unverified. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1unverified

Licence: 1 of the 1 sample is pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from cf020031308/LinkDist. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

cf020031308/LinkDist official repository 1 sample, 0 ran · 1 unverified
optimize in 11 papers ogbn.py unverified no licence file found · pointer only · eaab7909ebfb427e · report

Tasks

Node ClassificationNode Property Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Classification Amazon Computers LinkDist Accuracy 89.49% #9 of 12 Archive leaderboard report
Node Classification Amazon Computers LinkDistMLP Accuracy 89.44% #10 of 12 Archive leaderboard report
Node Classification Amazon Computers CoLinkDist Accuracy 89.42% #11 of 12 Archive leaderboard report
Node Classification Amazon Computers CoLinkDistMLP Accuracy 88.85% #12 of 12 Archive leaderboard report
Node Classification Amazon Photo CoLinkDist Accuracy 94.36% #8 of 11 Archive leaderboard report
Node Classification Amazon Photo CoLinkDistMLP Accuracy 94.12% #9 of 11 Archive leaderboard report
Node Classification Amazon Photo LinkDistMLP Accuracy 93.83% #10 of 11 Archive leaderboard report
Node Classification Amazon Photo LinkDist Accuracy 93.75% #11 of 11 Archive leaderboard report
Node Classification CiteSeer with Public Split: fixed 20 nodes per class CoLinkDistMLP Accuracy 70.96% #29 of 40 Archive leaderboard report
Node Classification CiteSeer with Public Split: fixed 20 nodes per class CoLinkDist Accuracy 70.79% #30 of 40 Archive leaderboard report
Node Classification CiteSeer with Public Split: fixed 20 nodes per class LinkDist Accuracy 70.27% #31 of 40 Archive leaderboard report
Node Classification CiteSeer with Public Split: fixed 20 nodes per class LinkDistMLP Accuracy 70.26% #32 of 40 Archive leaderboard report
Node Classification Citeseer CoLinkDist Accuracy 75.79% #17 of 71 Archive leaderboard report
Node Classification Citeseer CoLinkDistMLP Accuracy 75.77% #18 of 71 Archive leaderboard report
Node Classification Citeseer LinkDistMLP Accuracy 75.25% #22 of 71 Archive leaderboard report
Node Classification Citeseer LinkDist Accuracy 74.72% #24 of 71 Archive leaderboard report
Node Classification Coauthor CS CoLinkDist Accuracy 95.80% #6 of 24 Archive leaderboard report
Node Classification Coauthor CS CoLinkDistMLP Accuracy 95.74% #7 of 24 Archive leaderboard report
Node Classification Coauthor CS LinkDistMLP Accuracy 95.68% #9 of 24 Archive leaderboard report
Node Classification Coauthor CS LinkDist Accuracy 95.66% #11 of 24 Archive leaderboard report
Node Classification Coauthor Physics CoLinkDist Accuracy 97.05% #5 of 14 Archive leaderboard report
Node Classification Coauthor Physics LinkDistMLP Accuracy 96.91% #8 of 14 Archive leaderboard report
Node Classification Coauthor Physics CoLinkDistMLP Accuracy 96.87% #10 of 14 Archive leaderboard report
Node Classification Coauthor Physics LinkDist Accuracy 96.87% #11 of 14 Archive leaderboard report
Node Classification Cora LinkDist Accuracy 88.24% #10 of 73 Archive leaderboard report
Node Classification Cora CoLinkDist Accuracy 87.89% #12 of 73 Archive leaderboard report
Node Classification Cora LinkDistMLP Accuracy 87.58% #14 of 73 Archive leaderboard report
Node Classification Cora CoLinkDistMLP Accuracy 87.54% #15 of 73 Archive leaderboard report
Node Classification Cora Full CoLinkDist Accuracy 70.32% #2 of 5 Archive leaderboard report
Node Classification Cora Full LinkDist Accuracy 69.87% #3 of 5 Archive leaderboard report
Node Classification Cora Full CoLinkDistMLP Accuracy 69.83% #4 of 5 Archive leaderboard report
Node Classification Cora Full LinkDistMLP Accuracy 69.53% #5 of 5 Archive leaderboard report
Node Classification Cora Full with Public Split CoLinkDist Accuracy 57.05% #1 of 4 Archive leaderboard report
Node Classification Cora Full with Public Split LinkDist Accuracy 55.87% #2 of 4 Archive leaderboard report
Node Classification Cora Full with Public Split CoLinkDistMLP Accuracy 53.43% #3 of 4 Archive leaderboard report
Node Classification Cora Full with Public Split LinkDistMLP Accuracy 51.78% #4 of 4 Archive leaderboard report
Node Classification Cora with Public Split: fixed 20 nodes per class CoLinkDist Accuracy 81.39% #26 of 36 Archive leaderboard report
Node Classification Cora with Public Split: fixed 20 nodes per class CoLinkDistMLP Accuracy 81.19% #27 of 36 Archive leaderboard report
Node Classification Cora with Public Split: fixed 20 nodes per class LinkDist Accuracy 81.05% #28 of 36 Archive leaderboard report
Node Classification Cora with Public Split: fixed 20 nodes per class LinkDistMLP Accuracy 80.79% #29 of 36 Archive leaderboard report
Node Classification PubMed with Public Split: fixed 20 nodes per class CoLinkDist Accuracy 75.64% #33 of 37 Archive leaderboard report
Node Classification PubMed with Public Split: fixed 20 nodes per class CoLinkDistMLP Accuracy 75.41% #34 of 37 Archive leaderboard report
Node Classification PubMed with Public Split: fixed 20 nodes per class LinkDist Accuracy 74.06% #35 of 37 Archive leaderboard report
Node Classification PubMed with Public Split: fixed 20 nodes per class LinkDistMLP Accuracy 72.41% #36 of 37 Archive leaderboard report
Node Classification Pubmed CoLinkDist Accuracy 89.58% #9 of 70 Archive leaderboard report
Node Classification Pubmed CoLinkDistMLP Accuracy 89.53% #10 of 70 Archive leaderboard report
Node Classification Pubmed LinkDist Accuracy 88.86% #14 of 70 Archive leaderboard report
Node Classification Pubmed LinkDistMLP Accuracy 88.79% #15 of 70 Archive leaderboard report
Node Property Prediction ogbn-arxiv CoLinkDistMLP Ext. data No #83 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv CoLinkDistMLP Number of params 120912 #83 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv CoLinkDistMLP Test Accuracy 0.5638 ± 0.0016 #83 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv CoLinkDistMLP Validation Accuracy 0.5807 ± 0.0011 #83 of 86 Archive leaderboard report
Node Property Prediction ogbn-mag CoLinkDistMLP Ext. data No #38 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag CoLinkDistMLP Number of params 278202 #38 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag CoLinkDistMLP Test Accuracy 0.2761 ± 0.0018 #38 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag CoLinkDistMLP Validation Accuracy 0.2646 ± 0.0013 #38 of 39 Archive leaderboard report
Node Property Prediction ogbn-products CoLinkDistMLP Ext. data No #62 of 64 Archive leaderboard report
Node Property Prediction ogbn-products CoLinkDistMLP Number of params 115806 #62 of 64 Archive leaderboard report
Node Property Prediction ogbn-products CoLinkDistMLP Test Accuracy 0.6259 ± 0.0010 #62 of 64 Archive leaderboard report
Node Property Prediction ogbn-products CoLinkDistMLP Validation Accuracy 0.7721 ± 0.0015 #62 of 64 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections