Papers › SCR: Training Graph Neural Networks with Consistency Regularization

SCR: Training Graph Neural Networks with Consistency Regularization

8 Dec 2021arXiv:2112.04319archive 2025-07-28

Chenhui Zhang, Yufei He, Yukuo Cen, Zhenyu Hou, Wenzheng Feng, Yuxiao Dong, Xu Cheng, Hongyun Cai, Feng He, Jie Tang

We present the SCR framework for enhancing the training of graph neural networks (GNNs) with consistency regularization. Regularization is a set of strategies used in Machine Learning to reduce overfitting and improve the generalization ability. However, it is unclear how to best design the generalization strategies in GNNs, as it works in a semi-supervised setting for graph data. The major challenge lies in how to efficiently balance the trade-off between the error from the labeled data and that from the unlabeled data. SCR is a simple yet general framework in which we introduce two strategies of consistency regularization to address the challenge above. One is to minimize the disagreements among the perturbed predictions by different versions of a GNN model. The other is to leverage the Mean Teacher paradigm to estimate a consistency loss between teacher and student models instead of the disagreement of the predictions. We conducted experiments on three large-scale node classification datasets in the Open Graph Benchmark (OGB). Experimental results demonstrate that the proposed SCR framework is a general one that can enhance various GNNs to achieve better performance. Finally, SCR has been the top-1 entry on all three OGB leaderboards as of this submission.

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 3 code samples: 2 from 1 repository linked to this paper, 1 identical to code harvested from another paper's repository. Across all 3: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 2 samples from 1 repository, 1 ran (1 with no instrument failure; 0 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

thudm/crgnn officialmentioned in papermentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
thudm/scr officialmentioned in paperpytorchMITSyntology: not harvested report
THUDM/CRGNN mentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
THUDM/SCR pytorchMITSyntology: not harvested 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

3 samples harvested: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. 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.

1ran · honoured contract
1ran
1unverified

Licence: 1 of the 3 samples 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 thudm/crgnn. 1 sample is listed last, in a group without a repository. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “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.

thudm/crgnn official repository 2 samples, 1 ran (0 instrument-failed) · 1 unverified
read_relation_subsets ogbn-mag/heteo_data.py ran MIT (permissive) · 198da4ebc08d7f6a · report fixture: signature fname:str/0/str · returned list · level recorded 2026-09-24
load_data in 700 papers ogbn-mag/heteo_data.py unverified MIT (permissive) · 7fe187a46e4207d3 · report
Repository not located 1 sample, 1 ran (0 instrument-failed) · 0 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
get_n_params in 21 papers repository not located ran · honoured contract licence of this copy not recorded · 608920c78f1ae5e9 · report fixture: signature model:callable/0/? · returned int · level recorded 2026-09-22

Tasks

Node ClassificationNode Property Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Property Prediction ogbn-mag NARS-GAMLP+RLU+SCR Ext. data No #12 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+RLU+SCR Number of params 6734882 #12 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+RLU+SCR Test Accuracy 0.5631 ± 0.0021 #12 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+RLU+SCR Validation Accuracy 0.5734 ± 0.0035 #12 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR-m Ext. data No #15 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR-m Number of params 6734882 #15 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR-m Test Accuracy 0.5451 ± 0.0019 #15 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR-m Validation Accuracy 0.5590 ± 0.0028 #15 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR Ext. data No #17 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR Number of params 6734882 #17 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR Test Accuracy 0.5432 ± 0.0018 #17 of 39 Archive leaderboard report
Node Property Prediction ogbn-mag NARS-GAMLP+SCR Validation Accuracy 0.5654 ± 0.0021 #17 of 39 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+RLU+SCR Ext. data No #3 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+RLU+SCR Number of params 67560875 #3 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+RLU+SCR Test Accuracy 0.6842 ± 0.0015 #3 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+RLU+SCR Validation Accuracy 0.7188 ± 0.0007 #3 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR-m Ext. data No #6 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR-m Number of params 67560875 #6 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR-m Test Accuracy 0.6816 ± 0.0012 #6 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR-m Validation Accuracy 0.7186 ± 0.0008 #6 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR Ext. data No #7 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR Number of params 67560875 #7 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR Test Accuracy 0.6814 ± 0.0008 #7 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M GAMLP+SCR Validation Accuracy 0.7190 ± 0.0007 #7 of 20 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR+C&S Ext. data Yes #7 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR+C&S Number of params 1154654 #7 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR+C&S Test Accuracy 0.8680 ± 0.0007 #7 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR+C&S Validation Accuracy 0.9357 ± 0.0004 #7 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR+C&S Ext. data Yes #8 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR+C&S Number of params 1154654 #8 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR+C&S Test Accuracy 0.8673 ± 0.0008 #8 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR+C&S Validation Accuracy 0.9387 ± 0.0002 #8 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR Ext. data Yes #9 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR Number of params 1154654 #9 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR Test Accuracy 0.8667 ± 0.0009 #9 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+SCR Validation Accuracy 0.9364 ± 0.0005 #9 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR Ext. data Yes #10 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR Number of params 1154654 #10 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR Test Accuracy 0.8651 ± 0.0009 #10 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+SAGN+MCR Validation Accuracy 0.9389 ± 0.0002 #10 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+GAMLP+MCR Ext. data Yes #14 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+GAMLP+MCR Number of params 2144151 #14 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+GAMLP+MCR Test Accuracy 0.8591 ± 0.0008 #14 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GIANT-XRT+GAMLP+MCR Validation Accuracy 0.9402 ± 0.0004 #14 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR+C&S Ext. data No #15 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR+C&S Number of params 3335831 #15 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR+C&S Test Accuracy 0.8520 ± 0.0008 #15 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR+C&S Validation Accuracy 0.9304 ± 0.0005 #15 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR Ext. data No #16 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR Number of params 3335831 #16 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR Test Accuracy 0.8505 ± 0.0009 #16 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+RLU+SCR Validation Accuracy 0.9292 ± 0.0005 #16 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+MCR Ext. data No #19 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+MCR Number of params 3335831 #19 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+MCR Test Accuracy 0.8462 ± 0.0003 #19 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GAMLP+MCR Validation Accuracy 0.9319 ± 0.0003 #19 of 64 Archive leaderboard report
Node Property Prediction ogbn-products SAGN+MCR Ext. data No #22 of 64 Archive leaderboard report
Node Property Prediction ogbn-products SAGN+MCR Number of params 2179678 #22 of 64 Archive leaderboard report
Node Property Prediction ogbn-products SAGN+MCR Test Accuracy 0.8441 ± 0.0005 #22 of 64 Archive leaderboard report
Node Property Prediction ogbn-products SAGN+MCR Validation Accuracy 0.9325 ± 0.0004 #22 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