Papers › Pure Transformers are Powerful Graph Learners

Pure Transformers are Powerful Graph Learners

6 Jul 2022arXiv:2207.02505archive 2025-07-28

Jinwoo Kim, Tien Dat Nguyen, Seonwoo Min, Sungjun Cho, Moontae Lee, Honglak Lee, Seunghoon Hong

We show that standard Transformers without graph-specific modifications can lead to promising results in graph learning both in theory and practice. Given a graph, we simply treat all nodes and edges as independent tokens, augment them with token embeddings, and feed them to a Transformer. With an appropriate choice of token embeddings, we prove that this approach is theoretically at least as expressive as an invariant graph network (2-IGN) composed of equivariant linear layers, which is already more expressive than all message-passing Graph Neural Networks (GNN). When trained on a large-scale graph dataset (PCQM4Mv2), our method coined Tokenized Graph Transformer (TokenGT) achieves significantly better results compared to GNN baselines and competitive results compared to Transformer variants with sophisticated graph-specific inductive bias. Our implementation is available at https://github.com/jw9730/tokengt.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2207.02505")

Code

Syntology Ran 7 of 7 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 2 ran · violated contract; 5 ran with no contract checked.

By repository: official repository: 7 samples from 1 repository, 7 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

jw9730/tokengt officialmentioned in papermentioned on GitHubpytorchMIT report
luis-mueller/wl-transformers mentioned on GitHubpytorch 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

7 samples harvested; 7 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · violated contract
5ran

Licence: 7 of the 7 samples are 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 jw9730/tokengt. “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 “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

default jw9730/tokengt/large-scale-regression/tokengt/modules/performer_pytorch.py official repository ran · violated contract fingerprinted MIT recorded; this copy not marked cleared · pointer only · 60fff7c3c400d7ff · report
drop_path jw9730/tokengt/large-scale-regression/tokengt/modules/droppath.py official repository ran MIT recorded; this copy not marked cleared · pointer only · 49eae9f4f9e2a885 · report
empty jw9730/tokengt/large-scale-regression/tokengt/modules/performer_pytorch.py official repository ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 19925529d217754e · report
exists jw9730/tokengt/large-scale-regression/tokengt/modules/performer_pytorch.py official repository ran · violated contract MIT recorded; this copy not marked cleared · pointer only · aa5486a3650902d8 · report
gaussian_orthogonal_random_matrix jw9730/tokengt/large-scale-regression/tokengt/modules/orf.py official repository ran MIT recorded; this copy not marked cleared · pointer only · f13c4bce34f13104 · report
orthogonal_matrix_chunk jw9730/tokengt/large-scale-regression/tokengt/modules/orf.py official repository ran MIT recorded; this copy not marked cleared · pointer only · a02dd771934e8662 · report
orthogonal_matrix_chunk_batched jw9730/tokengt/large-scale-regression/tokengt/modules/orf.py official repository ran MIT recorded; this copy not marked cleared · pointer only · 23f8c3620122768a · report

Tasks

Graph ClassificationGraph LearningGraph RegressionInductive BiasMolecular Property Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification D&D TokenGT Accuracy 73.950±3.361 #48 of 53 Archive leaderboard report
Graph Classification IMDb-B TokenGT Accuracy 80.250±3.304 #6 of 51 Archive leaderboard report
Graph Classification NCI1 TokenGT Accuracy 76.740±2.054 #46 of 69 Archive leaderboard report
Graph Classification NCI109 TokenGT Accuracy 72.077±1.883 #33 of 38 Archive leaderboard report
Graph Regression ESR2 TokenGT R2 0.641±0.000 #8 of 9 Archive leaderboard report
Graph Regression ESR2 TokenGT RMSE 0.529±0.641 #8 of 9 Archive leaderboard report
Graph Regression F2 TokenGT R2 0.872±0.000 #8 of 9 Archive leaderboard report
Graph Regression F2 TokenGT RMSE 0.363±0.872 #8 of 9 Archive leaderboard report
Graph Regression KIT TokenGT R2 0.800±0.000 #8 of 9 Archive leaderboard report
Graph Regression KIT TokenGT RMSE 0.486±0.800 #8 of 9 Archive leaderboard report
Graph Regression Lipophilicity TokenGT R2 0.545±0.024 #16 of 23 Archive leaderboard report
Graph Regression Lipophilicity TokenGT RMSE 0.852±0.023 #16 of 23 Archive leaderboard report
Graph Regression PARP1 TokenGT R2 0.907±0.000 #8 of 9 Archive leaderboard report
Graph Regression PARP1 TokenGT RMSE 0.383±0.907 #8 of 9 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC TokenGT Test MAE 0.0919 #17 of 20 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC TokenGT Validation MAE 0.0910 #17 of 20 Archive leaderboard report
Graph Regression PGR TokenGT R2 0.684±0.000 #5 of 9 Archive leaderboard report
Graph Regression PGR TokenGT RMSE 0.543±0.684 #5 of 9 Archive leaderboard report
Graph Regression Peptides-struct TokenGT MAE 0.2489±0.0013 #21 of 39 Archive leaderboard report
Graph Regression ZINC-full TokenGT Test MAE 0.047±0.010 #13 of 19 Archive leaderboard report
Molecular Property Prediction ESOL TokenGT R2 0.892±0.036 #12 of 20 Archive leaderboard report
Molecular Property Prediction ESOL TokenGT RMSE 0.667±0.103 #12 of 20 Archive leaderboard report
Molecular Property Prediction FreeSolv TokenGT R2 0.930±0.018 #8 of 22 Archive leaderboard report
Molecular Property Prediction FreeSolv TokenGT RMSE 1.038±0.125 #8 of 22 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 ConnectionsDropoutGraph 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