Papers › Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification

Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification

13 Jun 2024arXiv:2406.08993archive 2025-07-28

Yuankai Luo, Lei Shi, Xiao-Ming Wu

Graph Transformers (GTs) have recently emerged as popular alternatives to traditional message-passing Graph Neural Networks (GNNs), due to their theoretically superior expressiveness and impressive performance reported on standard node classification benchmarks, often significantly outperforming GNNs. In this paper, we conduct a thorough empirical analysis to reevaluate the performance of three classic GNN models (GCN, GAT, and GraphSAGE) against GTs. Our findings suggest that the previously reported superiority of GTs may have been overstated due to suboptimal hyperparameter configurations in GNNs. Remarkably, with slight hyperparameter tuning, these classic GNN models achieve state-of-the-art performance, matching or even exceeding that of recent GTs across 17 out of the 18 diverse datasets examined. Additionally, we conduct detailed ablation studies to investigate the influence of various GNN configurations, such as normalization, dropout, residual connections, and network depth, on node classification performance. Our study aims to promote a higher standard of empirical rigor in the field of graph machine learning, encouraging more accurate comparisons and evaluations of model capabilities.

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 9 code samples harvested from 1 repository linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 9 samples from 1 repository, 5 ran (3 with no instrument failure; 2 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.

Levels written between 2026-09-24 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

LUOyk1999/tunedGNN officialmentioned in papermentioned on GitHubpytorchMITSyntology: 9 samples, 5 ran (2 instrument-failed) · 4 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

9 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
1ran; the contract check could not run
3ran
4unverified

Licence: 3 of the 9 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from LUOyk1999/tunedGNN. “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.

LUOyk1999/tunedGNN official repository 9 samples, 5 ran (2 instrument-failed) · 4 unverified
class_rand_splits in 5 papers medium_graph/data_utils.py ran MIT (permissive) here; the stored copy is pointer only · 58baf2e0e8e1755c · report fixture: signature label:Tensor/2/int, label_num_per_class:int/0/int, valid_num:int/0/int, test_num:int/0/int · returned dict · level recorded 2026-09-26
eval_acc in 15 papers large_graph/data_utils.py ran · our draft was wrong MIT (permissive) · 319c36c17da75fed · report fixture: signature y_true:Tensor/2/int, y_pred:Tensor/3/float · returned float · level recorded 2026-09-24
load_fixed_splits in 12 papers medium_graph/data_utils.py ran MIT (permissive) here; the stored copy is pointer only · 025e94d610e74ffd · report fixture: signature data_dir:str/0/str, dataset:NoneType/0/?, name:str/0/str · returned list · level recorded 2026-09-26
load_fixed_splits in 12 papers large_graph/data_utils.py ran MIT (permissive) · 5b7702a29f6dafaf · report fixture: signature data_dir:str/0/str, dataset:NoneType/0/?, name:str/0/str · returned list · level recorded 2026-09-24
rand_train_test_idx in 18 papers medium_graph/data_utils.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only · f54bada0dd9396ba · report fixture: signature label:Tensor/1/int, train_prop:float/0/float, valid_prop:float/0/float, ignore_negative:bool/0/bool · returned tuple · level recorded 2026-09-26
eval_f1 large_graph/data_utils.py unverified MIT (permissive) · 39f7b3a8e45be81b · report
load_hetero_dataset large_graph/dataset.py unverified MIT (permissive) · 2ca80bf081304f24 · report
load_model in 479 papers large_graph/logger.py unverified MIT (permissive) · 54fd4512b8397f45 · report
load_wikics_dataset large_graph/dataset.py unverified MIT (permissive) · a411b546b0451f0e · report

Tasks

Node ClassificationNode Property Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Classification Amazon Computers GAT Accuracy 94.09±0.37 #1 of 12 Archive leaderboard report
Node Classification Amazon Computers GCN Accuracy 93.99±0.12 #2 of 12 Archive leaderboard report
Node Classification Amazon Computers GraphSAGE Accuracy 93.25±0.14 #3 of 12 Archive leaderboard report
Node Classification Amazon Photo GraphSAGE Accuracy 96.78 ± 0.23 #1 of 11 Archive leaderboard report
Node Classification Amazon Photo GAT Accuracy 96.60 ± 0.33 #2 of 11 Archive leaderboard report
Node Classification Amazon Photo GCN Accuracy 96.10 ± 0.46 #3 of 11 Archive leaderboard report
Node Classification CiteSeer with Public Split: fixed 20 nodes per class GCN Accuracy 73.14± 0.67 #20 of 40 Archive leaderboard report
Node Classification Coauthor CS GraphSAGE Accuracy 96.38±0.11 #3 of 24 Archive leaderboard report
Node Classification Coauthor Physics GCN Accuracy 97.46 ± 0.10 #3 of 14 Archive leaderboard report
Node Classification Cora with Public Split: fixed 20 nodes per class GCN Accuracy 85.1 ± 0.7 #6 of 36 Archive leaderboard report
Node Classification PubMed with Public Split: fixed 20 nodes per class GCN Accuracy 81.12 ± 0.52 #8 of 37 Archive leaderboard report
Node Classification amazon-ratings GAT Accuracy (%) 55.54 ± 0.51 #1 of 4 Archive leaderboard report
Node Classification amazon-ratings GraphSAGE Accuracy (%) 55.40 ± 0.21 #2 of 4 Archive leaderboard report
Node Classification amazon-ratings GCN Accuracy (%) 53.80 ± 0.60 #4 of 4 Archive leaderboard report
Node Classification minesweeper GCN AUCROC 97.86 ± 0.24 #1 of 4 Archive leaderboard report
Node Classification minesweeper GraphSAGE AUCROC 97.77 ± 0.62 #2 of 4 Archive leaderboard report
Node Classification minesweeper GAT AUCROC 97.73 ± 0.73 #3 of 4 Archive leaderboard report
Node Classification pokec GCN Accuracy 86.33 ± 0.17 #2 of 7 Archive leaderboard report
Node Classification questions GCN AUCROC 79.02 ± 0.60 #2 of 3 Archive leaderboard report
Node Classification roman-empire GCN Accuracy (% ) 91.27±0.20 #4 of 7 Archive leaderboard report
Node Property Prediction ogbn-arxiv GCN Ext. data No #43 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GCN Number of params 1463336 #43 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GCN Test Accuracy 0.7360 ± 0.0018 #43 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GCN Validation Accuracy 0.7447 ± 0.0014 #43 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GraphSAGE Ext. data No #53 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GraphSAGE Number of params 1727272 #53 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GraphSAGE Test Accuracy 0.7295 ± 0.0031 #53 of 86 Archive leaderboard report
Node Property Prediction ogbn-arxiv GraphSAGE Validation Accuracy 0.7397 ± 0.0015 #53 of 86 Archive leaderboard report
Node Property Prediction ogbn-products GraphSAGE Ext. data No #26 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GraphSAGE Number of params 433047 #26 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GraphSAGE Test Accuracy 0.8389 ± 0.0036 #26 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GraphSAGE Validation Accuracy 0.9242 ± 0.0029 #26 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GCN Ext. data No #34 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GCN Number of params 233047 #34 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GCN Test Accuracy 0.8233 ± 0.0019 #34 of 64 Archive leaderboard report
Node Property Prediction ogbn-products GCN Validation Accuracy 0.9224 ± 0.0036 #34 of 64 Archive leaderboard report
Node Property Prediction ogbn-proteins GAT Ext. data No #15 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GAT Number of params 2943472 #15 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GAT Test ROC-AUC 0.8501 ± 0.0046 #15 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GAT Validation ROC-AUC 0.9067 ± 0.0043 #15 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GraphSAGE Ext. data No #19 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GraphSAGE Number of params 2444896 #19 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GraphSAGE Test ROC-AUC 0.8221 ± 0.0032 #19 of 26 Archive leaderboard report
Node Property Prediction ogbn-proteins GraphSAGE Validation ROC-AUC 0.8831 ± 0.0044 #19 of 26 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.

Methods

GAT

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