Papers › Where Did the Gap Go? Reassessing the Long-Range Graph Benchmark

Where Did the Gap Go? Reassessing the Long-Range Graph Benchmark

1 Sep 2023arXiv:2309.00367archive 2025-07-28

Jan Tönshoff, Martin Ritzert, Eran Rosenbluth, Martin Grohe

The recent Long-Range Graph Benchmark (LRGB, Dwivedi et al. 2022) introduced a set of graph learning tasks strongly dependent on long-range interaction between vertices. Empirical evidence suggests that on these tasks Graph Transformers significantly outperform Message Passing GNNs (MPGNNs). In this paper, we carefully reevaluate multiple MPGNN baselines as well as the Graph Transformer GPS (Rampášek et al. 2022) on LRGB. Through a rigorous empirical analysis, we demonstrate that the reported performance gap is overestimated due to suboptimal hyperparameter choices. It is noteworthy that across multiple datasets the performance gap completely vanishes after basic hyperparameter optimization. In addition, we discuss the impact of lacking feature normalization for LRGB's vision datasets and highlight a spurious implementation of LRGB's link prediction metric. The principal aim of our paper is to establish a higher standard of empirical rigor within the graph machine learning community.

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 12 code samples harvested from 2 repositories linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 4 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 10 samples from 1 repository, 7 ran (5 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 2 samples from 1 repository, 1 ran (1 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.

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 8 samples that ran have no recorded date.

toenshoff/lrgb officialmentioned in paperpytorchMITSyntology: 10 samples, 7 ran (2 instrument-failed) · 3 unverified report
Fedzbar/laser-release mentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (0 instrument-failed) · 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

12 samples harvested: 8 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 4 violated, 2 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.

4ran · violated contract
2ran · our draft was wrong
2ran
4unverified

Licence: 6 of the 12 samples are pointer only, meaning Syntology does not serve that copy's text. For 6 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

toenshoff/lrgb official repository 10 samples, 7 ran (2 instrument-failed) · 3 unverified
default in 430 papers graphgps/layer/performer_layer.py ran · violated contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 60fff7c3c400d7ff · report fixture: signature val:int/0/int, d:int/0/int · returned int · level recorded, date not recorded
empty in 9 papers graphgps/layer/performer_layer.py ran · our draft was wrong fingerprinted MIT (permissive) · 19925529d217754e · report fixture: signature tensor:Tensor/2/float · returned bool · level recorded 2026-09-24
exists in 528 papers
census: 5 implementations from 103 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 1, kind int, dtype int64
graphgps/layer/performer_layer.py ran · violated contract MIT (permissive) here; the stored copy is pointer only · aa5486a3650902d8 · report fixture: signature val:Tensor/1/int · returned bool · level recorded, date not recorded
get_final_pretrained_ckpt in 28 papers graphgps/finetuning.py ran · our draft was wrong MIT (permissive) · 1bb331bf70a78136 · report fixture: signature ckpt_dir:str/0/str · returned str · level recorded 2026-09-24
is_seed in 16 papers graphgps/agg_runs.py ran · violated contract MIT (permissive) here; the stored copy is pointer only · c98a702be0660ac9 · report fixture: signature s:str/0/str · returned bool · level recorded 2026-09-25
is_split in 17 papers graphgps/agg_runs.py ran · violated contract MIT (permissive) here; the stored copy is pointer only · 438635c08adfe2aa · report fixture: signature s:str/0/str · returned bool · level recorded 2026-09-25
join_list in 19 papers graphgps/agg_runs.py ran fingerprinted MIT (permissive) here; the stored copy is pointer only · 40e98e1ecbf39f34 · report fixture: signature l1:list/1/Tensor, l2:list/1/Tensor · returned list · level recorded 2026-09-25
apply_chunking_to_forward in 25 papers graphgps/layer/bigbird_layer.py unverified MIT (permissive) here; the stored copy is pointer only · ae56bcfaae96da3c · report
init_model_from_pretrained in 28 papers graphgps/finetuning.py unverified MIT (permissive) · 3db4474abe13fdee · report
load_pretrained_model_cfg in 27 papers graphgps/finetuning.py unverified MIT (permissive) · d796cdea5889c847 · report
Fedzbar/laser-release community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
init_model_from_pretrained in 28 papers laser/finetuning.py ran MIT (permissive) · 8653a762bb268420 · report fixture: signature model:callable/0/?, pretrained_dir:str/0/str, freeze_pretrained:bool/0/bool · returned SimpleModel · level recorded 2026-09-24
load_pretrained_model_cfg in 27 papers laser/finetuning.py unverified MIT (permissive) · 995500eb98f90144 · report

Tasks

Graph ClassificationGraph LearningGraph RegressionHyperparameter OptimizationLink PredictionNode Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification Peptides-func GCN-tuned AP 0.6860±0.0050 #20 of 44 Archive leaderboard report
Graph Classification Peptides-func GatedGCN-tuned AP 0.6765±0.0047 #23 of 44 Archive leaderboard report
Graph Classification Peptides-func GINE-tuned AP 0.6621±0.0067 #27 of 44 Archive leaderboard report
Graph Classification Peptides-func GPS-tuned AP 0.6534±0.0091 #31 of 44 Archive leaderboard report
Graph Regression Peptides-struct GCN-tuned MAE 0.2460±0.0007 #11 of 39 Archive leaderboard report
Graph Regression Peptides-struct GINE-tuned MAE 0.2473±0.0017 #14 of 39 Archive leaderboard report
Graph Regression Peptides-struct GatedGCN-tuned MAE 0.2477±0.0009 #16 of 39 Archive leaderboard report
Graph Regression Peptides-struct GPS-tuned MAE 0.2509±0.0014 #24 of 39 Archive leaderboard report
Link Prediction PCQM-Contact GINE-tuned MRR 0.3509±0.0006 #12 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GINE-tuned MRR-ext-filtered 0.4617±0.0005 #12 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GPS-tuned MRR 0.3498±0.0005 #13 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GPS-tuned MRR-ext-filtered 0.4703±0.0014 #13 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GatedGCN-tuned MRR 0.3495±0.0010 #14 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GatedGCN-tuned MRR-ext-filtered 0.4670±0.0004 #14 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GCN-tuned MRR 0.3424±0.0007 #16 of 18 Archive leaderboard report
Link Prediction PCQM-Contact GCN-tuned MRR-ext-filtered 0.4526±0.0006 #16 of 18 Archive leaderboard report
Node Classification COCO-SP GPS-tuned macro F1 0.3884±0.0055 #2 of 19 Archive leaderboard report
Node Classification COCO-SP GatedGCN-tuned macro F1 0.2922±0.0018 #8 of 19 Archive leaderboard report
Node Classification COCO-SP GINE-tuned macro F1 0.2125±0.0009 #14 of 19 Archive leaderboard report
Node Classification COCO-SP GCN-tuned macro F1 0.1338±0.0007 #18 of 19 Archive leaderboard report
Node Classification PascalVOC-SP GPS-tuned macro F1 0.4440±0.0065 #3 of 21 Archive leaderboard report
Node Classification PascalVOC-SP GatedGCN-tuned macro F1 0.3880±0.0040 #7 of 21 Archive leaderboard report
Node Classification PascalVOC-SP GINE-tuned macro F1 0.2718±0.0054 #16 of 21 Archive leaderboard report
Node Classification PascalVOC-SP GCN-tuned macro F1 0.2078±0.0031 #18 of 21 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

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPSGraph TransformerLabel SmoothingLapEigenLaplacian PELayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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