{"url":"/sota/graph-regression-on-pcqm4mv2-lsc","task":{"name":"Graph Regression","url":"/task/graph-regression","note":null},"dataset":{"name":"PCQM4Mv2-LSC","url":"/dataset/pcqm4mv2-lsc"},"category":"Graphs","categories":["Graphs"],"category_note":null,"description":"The regression task is similar to graph classification but using different loss function and performance metric.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Validation MAE","Test MAE"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Validation MAE":"lower","Test MAE":"lower"}},"counts":{"rows":20,"rows_with_code":20,"rows_with_paper_page":20,"rows_dated":20,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"ESA (Edge set attention, no positional encodings)","metrics":{"Test MAE":"N/A","Validation MAE":"0.0235"},"uses_additional_data":false,"paper_date":"2024-02-16","paper":"/paper/masked-attention-is-all-you-need-for-graphs","paper_url":"https://arxiv.org/abs/2402.10793v2","paper_title":"An end-to-end attention-based approach for learning on graphs","code":"https://github.com/davidbuterez/edge-set-attention","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"EGT + Triangular Attention","metrics":{"Test MAE":"0.0683","Validation MAE":"0.0671"},"uses_additional_data":false,"paper_date":"2021-08-07","paper":"/paper/edge-augmented-graph-transformers-global-self","paper_url":"https://arxiv.org/abs/2108.03348v3","paper_title":"Global Self-Attention as a Replacement for Graph Convolution","code":"https://github.com/shamim-hussain/egt_pytorch","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"TGT-At","metrics":{"Test MAE":"0.0683","Validation MAE":"0.0671"},"uses_additional_data":false,"paper_date":"2024-02-07","paper":"/paper/triplet-interaction-improves-graph","paper_url":"https://arxiv.org/abs/2402.04538v2","paper_title":"Triplet Interaction Improves Graph Transformers: Accurate Molecular Graph Learning with Triplet Graph Transformers","code":"https://github.com/shamim-hussain/egt_pytorch","n_code_links":3,"syntology":{"n_ran":10,"n_unverified":3,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"Uni-Mol+","metrics":{"Test MAE":"0.0705","Validation MAE":"0.0693"},"uses_additional_data":false,"paper_date":"2023-03-16","paper":"/paper/highly-accurate-quantum-chemical-property","paper_url":"https://arxiv.org/abs/2303.16982v2","paper_title":"Highly Accurate Quantum Chemical Property Prediction with Uni-Mol+","code":"https://github.com/deepmodeling/Uni-Mol","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"Transformer-M","metrics":{"Test MAE":"0.0782","Validation MAE":"0.0772"},"uses_additional_data":false,"paper_date":"2022-10-04","paper":"/paper/one-transformer-can-understand-both-2d-3d","paper_url":"https://arxiv.org/abs/2210.01765v4","paper_title":"One Transformer Can Understand Both 2D & 3D Molecular Data","code":"https://github.com/lsj2408/Transformer-M","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"GPTrans-L","metrics":{"Test MAE":"0.0821","Validation MAE":"0.0809"},"uses_additional_data":false,"paper_date":"2023-05-19","paper":"/paper/graph-propagation-transformer-for-graph","paper_url":"https://arxiv.org/abs/2305.11424v3","paper_title":"Graph Propagation Transformer for Graph Representation Learning","code":"https://github.com/czczup/gptrans","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":4,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":7,"model":"TIGT","metrics":{"Validation MAE":"0.0826"},"uses_additional_data":false,"paper_date":"2024-02-03","paper":"/paper/topology-informed-graph-transformer","paper_url":"https://arxiv.org/abs/2402.02005v1","paper_title":"Topology-Informed Graph Transformer","code":"https://github.com/leemingo/tigt","n_code_links":2,"syntology":{"n_ran":14,"n_unverified":7,"n_samples":21,"n_pointer_only_licence":21}},{"rank_in_archive_order":8,"model":"GPTrans-T","metrics":{"Test MAE":"0.0842","Validation MAE":"0.0833"},"uses_additional_data":false,"paper_date":"2023-05-19","paper":"/paper/graph-propagation-transformer-for-graph","paper_url":"https://arxiv.org/abs/2305.11424v3","paper_title":"Graph Propagation Transformer for Graph Representation Learning","code":"https://github.com/czczup/gptrans","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":4,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":9,"model":"GPS","metrics":{"Test MAE":"0.0862","Validation MAE":"0.0852"},"uses_additional_data":false,"paper_date":"2022-05-25","paper":"/paper/recipe-for-a-general-powerful-scalable-graph","paper_url":"https://arxiv.org/abs/2205.12454v4","paper_title":"Recipe for a General, Powerful, Scalable Graph Transformer","code":"https://github.com/rampasek/GraphGPS","n_code_links":4,"syntology":{"n_ran":3,"n_unverified":18,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"EGT","metrics":{"Test MAE":"0.0862","Validation MAE":"0.0857"},"uses_additional_data":false,"paper_date":"2021-08-07","paper":"/paper/edge-augmented-graph-transformers-global-self","paper_url":"https://arxiv.org/abs/2108.03348v3","paper_title":"Global Self-Attention as a Replacement for Graph Convolution","code":"https://github.com/shamim-hussain/egt_pytorch","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"GRIT","metrics":{"Validation MAE":"0.0859"},"uses_additional_data":false,"paper_date":"2023-05-27","paper":"/paper/graph-inductive-biases-in-transformers","paper_url":"https://arxiv.org/abs/2305.17589v1","paper_title":"Graph Inductive Biases in Transformers without Message Passing","code":"https://github.com/liamma/grit","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":6,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"Graphormer + GFSA","metrics":{"Validation MAE":"0.0860"},"uses_additional_data":false,"paper_date":"2023-12-07","paper":"/paper/graph-convolutions-enrich-the-self-attention","paper_url":"https://arxiv.org/abs/2312.04234v5","paper_title":"Graph Convolutions Enrich the Self-Attention in Transformers!","code":"https://github.com/jeongwhanchoi/gfsa","n_code_links":1,"syntology":{"n_ran":19,"n_unverified":10,"n_samples":29,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"Graphormer","metrics":{"Test MAE":"-","Validation MAE":"0.0864"},"uses_additional_data":false,"paper_date":"2021-06-09","paper":"/paper/do-transformers-really-perform-bad-for-graph","paper_url":"https://arxiv.org/abs/2106.05234v5","paper_title":"Do Transformers Really Perform Bad for Graph Representation?","code":"https://github.com/microsoft/Graphormer","n_code_links":5,"syntology":null},{"rank_in_archive_order":14,"model":"EGT+SSA+Self-ensemble","metrics":{"Validation MAE":"0.0865"},"uses_additional_data":false,"paper_date":"2023-06-02","paper":"/paper/the-information-pathways-hypothesis","paper_url":"https://arxiv.org/abs/2306.01705v1","paper_title":"The Information Pathways Hypothesis: Transformers are Dynamic Self-Ensembles","code":"https://github.com/shamim-hussain/ssa","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"GRPE-Large","metrics":{"Test MAE":"0.0876","Validation MAE":"0.0867"},"uses_additional_data":false,"paper_date":"2022-01-30","paper":"/paper/graph-self-attention-for-learning-graph","paper_url":"https://arxiv.org/abs/2201.12787v3","paper_title":"GRPE: Relative Positional Encoding for Graph Transformer","code":"https://github.com/lenscloth/grpe","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"EGT+SSA","metrics":{"Validation MAE":"0.0876"},"uses_additional_data":false,"paper_date":"2023-06-02","paper":"/paper/the-information-pathways-hypothesis","paper_url":"https://arxiv.org/abs/2306.01705v1","paper_title":"The Information Pathways Hypothesis: Transformers are Dynamic Self-Ensembles","code":"https://github.com/shamim-hussain/ssa","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"TokenGT","metrics":{"Test MAE":"0.0919","Validation MAE":"0.0910"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/pure-transformers-are-powerful-graph-learners","paper_url":"https://arxiv.org/abs/2207.02505v2","paper_title":"Pure Transformers are Powerful Graph Learners","code":"https://github.com/jw9730/tokengt","n_code_links":2,"syntology":{"n_ran":7,"n_unverified":0,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":18,"model":"GIN","metrics":{"Test MAE":"0.1218","Validation MAE":"0.1195"},"uses_additional_data":false,"paper_date":"2018-10-01","paper":"/paper/how-powerful-are-graph-neural-networks","paper_url":"http://arxiv.org/abs/1810.00826v3","paper_title":"How Powerful are Graph Neural Networks?","code":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/gin","n_code_links":19,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":5}},{"rank_in_archive_order":19,"model":"GCN","metrics":{"Test MAE":"0.1398","Validation MAE":"0.1379"},"uses_additional_data":false,"paper_date":"2016-09-09","paper":"/paper/semi-supervised-classification-with-graph","paper_url":"http://arxiv.org/abs/1609.02907v4","paper_title":"Semi-Supervised Classification with Graph Convolutional Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/gcn","n_code_links":55,"syntology":{"n_ran":32,"n_unverified":26,"n_samples":58,"n_pointer_only_licence":22}},{"rank_in_archive_order":20,"model":"MLP-Fingerprint","metrics":{"Test MAE":"0.1760","Validation MAE":"0.1753"},"uses_additional_data":false,"paper_date":"2021-03-17","paper":"/paper/ogb-lsc-a-large-scale-challenge-for-machine","paper_url":"https://arxiv.org/abs/2103.09430v3","paper_title":"OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/ogb_lsc","n_code_links":6,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":15,"rows_with_any_sample_ran":12,"distinct_papers_with_graph_line":13,"distinct_papers_with_any_sample_ran":11,"samples_over_distinct_papers":{"n_ran":102,"n_unverified":89,"n_samples":191,"n_pointer_only_licence":63,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":106,"n_unverified":97,"n_samples":203,"n_pointer_only_licence":71,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}