{"url":"/sota/node-classification-on-squirrel-60-20-20","task":{"name":"Node Classification","url":"/task/node-classification","note":null},"dataset":{"name":"Squirrel (60%/20%/20% random splits)","url":"/dataset/squirrel-60-20-20-random-splits"},"category":"Graphs","categories":["Graphs"],"category_note":null,"description":"**Node Classification** is a machine learning task in graph-based data analysis, where the goal is to assign labels to nodes in a graph based on the properties of nodes and the relationships between them.\r\n\r\n**Node Classification** models aim to predict non-existing node properties (known as the target property) based on other node properties. Typical models used for node classification consists of a large family of graph neural networks. Model performance can be measured using benchmark datasets like [Cora](/dataset/cora), [Citeseer](/dataset/citeseer), and [Pubmed](/dataset/pubmed), among others, typically using Accuracy and F1.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Fast Graph Representation Learning With PyTorch Geometric](https://arxiv.org/pdf/1903.02428v3.pdf) )</span>","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":["1:1 Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"1:1 Accuracy":"higher"}},"counts":{"rows":37,"rows_with_code":36,"rows_with_paper_page":37,"rows_dated":37,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"GNNDLD","metrics":{"1:1 Accuracy":"77.72±0.84 "},"uses_additional_data":false,"paper_date":"2024-02-26","paper":"/paper/gnndld-graph-neural-network-with-directional","paper_url":"https://www.scitepress.org/Papers/2024/123214/","paper_title":"GNNDLD: Graph Neural Network with Directional Label Distribution","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"ACMII-GCN++","metrics":{"1:1 Accuracy":"69.98 ± 1.53"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"ACMII-GCN+","metrics":{"1:1 Accuracy":"69.81 ± 1.11"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"ACM-GCN+","metrics":{"1:1 Accuracy":"69.26 ± 1.11"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"ACM-GCN++","metrics":{"1:1 Accuracy":"68.56 ± 1.33"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"NFGNN","metrics":{"1:1 Accuracy":"58.9±0.35"},"uses_additional_data":false,"paper_date":"2022-12-07","paper":"/paper/node-oriented-spectral-filtering-for-graph","paper_url":"https://arxiv.org/abs/2212.03654v3","paper_title":"Node-oriented Spectral Filtering for Graph Neural Networks","code":"https://github.com/SsGood/NFGNN","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"ACM-Snowball-2","metrics":{"1:1 Accuracy":"55.97 ± 2.03"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"ACM-Snowball-3","metrics":{"1:1 Accuracy":"55.73 ± 2.39"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"ACMII-GCN","metrics":{"1:1 Accuracy":"54.53 ± 2.09"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"ACMII-Snowball-2","metrics":{"1:1 Accuracy":"53.48 ± 0.6"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"GCN+JK","metrics":{"1:1 Accuracy":"53.40 ± 1.90"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"ACMII-Snowball-3","metrics":{"1:1 Accuracy":"52.31 ± 1.57"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"GAT+JK","metrics":{"1:1 Accuracy":"52.28 ± 3.61"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"BernNet","metrics":{"1:1 Accuracy":"51.35 ± 0.73"},"uses_additional_data":false,"paper_date":"2021-06-21","paper":"/paper/bernnet-learning-arbitrary-graph-spectral","paper_url":"https://arxiv.org/abs/2106.10994v3","paper_title":"BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation","code":"https://github.com/ivam-he/BernNet","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":15,"model":"GPRGNN","metrics":{"1:1 Accuracy":"49.93 ± 0.53"},"uses_additional_data":false,"paper_date":"2020-06-14","paper":"/paper/joint-adaptive-feature-smoothing-and-topology","paper_url":"https://arxiv.org/abs/2006.07988v6","paper_title":"Adaptive Universal Generalized PageRank Graph Neural Network","code":"https://github.com/jianhao2016/GPRGNN","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":16,"model":"Snowball-3","metrics":{"1:1 Accuracy":"48.25 ± 0.94"},"uses_additional_data":false,"paper_date":"2019-06-05","paper":"/paper/break-the-ceiling-stronger-multi-scale-deep","paper_url":"https://arxiv.org/abs/1906.02174v3","paper_title":"Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks","code":"https://github.com/PwnerHarry/Stronger_GCN","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":8,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"Snowball-2","metrics":{"1:1 Accuracy":"47.88 ± 1.23"},"uses_additional_data":false,"paper_date":"2019-06-05","paper":"/paper/break-the-ceiling-stronger-multi-scale-deep","paper_url":"https://arxiv.org/abs/1906.02174v3","paper_title":"Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks","code":"https://github.com/PwnerHarry/Stronger_GCN","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":8,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"SGC-1","metrics":{"1:1 Accuracy":"47.62 ± 1.27"},"uses_additional_data":false,"paper_date":"2019-02-19","paper":"/paper/simplifying-graph-convolutional-networks","paper_url":"https://arxiv.org/abs/1902.07153v2","paper_title":"Simplifying Graph Convolutional Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/sgc","n_code_links":7,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"HH-GCN","metrics":{"1:1 Accuracy":"47.19 ± 1.21"},"uses_additional_data":false,"paper_date":"2023-08-17","paper":"/paper/half-hop-a-graph-upsampling-approach-for","paper_url":"https://arxiv.org/abs/2308.09198v1","paper_title":"Half-Hop: A graph upsampling approach for slowing down message passing","code":"https://github.com/nerdslab/halfhop","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"ACM-SGC-1","metrics":{"1:1 Accuracy":"46.4 ± 1.13"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"HH-GAT","metrics":{"1:1 Accuracy":"46.35 ± 1.86"},"uses_additional_data":false,"paper_date":"2023-08-17","paper":"/paper/half-hop-a-graph-upsampling-approach-for","paper_url":"https://arxiv.org/abs/2308.09198v1","paper_title":"Half-Hop: A graph upsampling approach for slowing down message passing","code":"https://github.com/nerdslab/halfhop","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"HH-GraphSAGE","metrics":{"1:1 Accuracy":"45.25 ± 1.52"},"uses_additional_data":false,"paper_date":"2023-08-17","paper":"/paper/half-hop-a-graph-upsampling-approach-for","paper_url":"https://arxiv.org/abs/2308.09198v1","paper_title":"Half-Hop: A graph upsampling approach for slowing down message passing","code":"https://github.com/nerdslab/halfhop","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"GCN","metrics":{"1:1 Accuracy":"44.76 ± 1.39"},"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":24,"model":"GAT","metrics":{"1:1 Accuracy":"42.72 ± 0.33"},"uses_additional_data":false,"paper_date":"2017-10-30","paper":"/paper/graph-attention-networks","paper_url":"http://arxiv.org/abs/1710.10903v3","paper_title":"Graph Attention Networks","code":"https://github.com/labmlai/annotated_deep_learning_paper_implementations","n_code_links":93,"syntology":{"n_ran":51,"n_unverified":55,"n_samples":106,"n_pointer_only_licence":43}},{"rank_in_archive_order":25,"model":"FAGCN","metrics":{"1:1 Accuracy":"42.24 ± 1.2"},"uses_additional_data":false,"paper_date":"2021-01-04","paper":"/paper/beyond-low-frequency-information-in-graph","paper_url":"https://arxiv.org/abs/2101.00797v1","paper_title":"Beyond Low-frequency Information in Graph Convolutional Networks","code":"https://github.com/bdy9527/FAGCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"GraphSAGE","metrics":{"1:1 Accuracy":"41.26 ± 0.26"},"uses_additional_data":false,"paper_date":"2017-06-07","paper":"/paper/inductive-representation-learning-on-large","paper_url":"http://arxiv.org/abs/1706.02216v4","paper_title":"Inductive Representation Learning on Large Graphs","code":"https://github.com/pyg-team/pytorch_geometric/blob/master/torch_geometric/nn/models/basic_gnn.py","n_code_links":20,"syntology":{"n_ran":3,"n_unverified":2,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":27,"model":"SGC-2","metrics":{"1:1 Accuracy":"41.25 ± 1.4"},"uses_additional_data":false,"paper_date":"2019-02-19","paper":"/paper/simplifying-graph-convolutional-networks","paper_url":"https://arxiv.org/abs/1902.07153v2","paper_title":"Simplifying Graph Convolutional Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/sgc","n_code_links":7,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":28,"model":"ACM-SGC-2","metrics":{"1:1 Accuracy":"40.91 ± 1.39"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"ACM-GCNII","metrics":{"1:1 Accuracy":"40.9 ± 1.58"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":30,"model":"GCNII","metrics":{"1:1 Accuracy":"38.81 ± 1.97"},"uses_additional_data":false,"paper_date":"2020-07-04","paper":"/paper/simple-and-deep-graph-convolutional-networks-1","paper_url":"https://arxiv.org/abs/2007.02133v1","paper_title":"Simple and Deep Graph Convolutional Networks","code":"https://github.com/chennnM/GCNII/tree/master/PyG/ogbn-arxiv","n_code_links":4,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":4}},{"rank_in_archive_order":31,"model":"ACM-GCNII*","metrics":{"1:1 Accuracy":"38.32 ± 1.5"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":32,"model":"GCNII*","metrics":{"1:1 Accuracy":"38.31 ±  1.3"},"uses_additional_data":false,"paper_date":"2020-07-04","paper":"/paper/simple-and-deep-graph-convolutional-networks-1","paper_url":"https://arxiv.org/abs/2007.02133v1","paper_title":"Simple and Deep Graph Convolutional Networks","code":"https://github.com/chennnM/GCNII/tree/master/PyG/ogbn-arxiv","n_code_links":4,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":4}},{"rank_in_archive_order":33,"model":"Geom-GCN*","metrics":{"1:1 Accuracy":"38.14"},"uses_additional_data":false,"paper_date":"2020-02-13","paper":"/paper/geom-gcn-geometric-graph-convolutional-1","paper_url":"https://arxiv.org/abs/2002.05287v2","paper_title":"Geom-GCN: Geometric Graph Convolutional Networks","code":"https://github.com/bingzhewei/geom-gcn","n_code_links":4,"syntology":{"n_ran":7,"n_unverified":0,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":34,"model":"APPNP","metrics":{"1:1 Accuracy":"34.77 ± 0.34"},"uses_additional_data":false,"paper_date":"2018-10-14","paper":"/paper/predict-then-propagate-graph-neural-networks","paper_url":"https://arxiv.org/abs/1810.05997v6","paper_title":"Predict then Propagate: Graph Neural Networks meet Personalized PageRank","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/appnp","n_code_links":5,"syntology":{"n_ran":2,"n_unverified":10,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":35,"model":"MLP-2","metrics":{"1:1 Accuracy":"31.28 ± 0.27"},"uses_additional_data":false,"paper_date":"2022-10-14","paper":"/paper/revisiting-heterophily-for-graph-neural","paper_url":"https://arxiv.org/abs/2210.07606v1","paper_title":"Revisiting Heterophily For Graph Neural Networks","code":"https://github.com/SitaoLuan/ACM-GNN","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":36,"model":"H2GCN","metrics":{"1:1 Accuracy":"30.39 ± 1.22"},"uses_additional_data":false,"paper_date":"2020-06-20","paper":"/paper/generalizing-graph-neural-networks-beyond","paper_url":"https://arxiv.org/abs/2006.11468v2","paper_title":"Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs","code":"https://github.com/GemsLab/H2GCN","n_code_links":4,"syntology":{"n_ran":13,"n_unverified":8,"n_samples":21,"n_pointer_only_licence":6}},{"rank_in_archive_order":37,"model":"MixHop","metrics":{"1:1 Accuracy":"24.55 ± 2.60"},"uses_additional_data":false,"paper_date":"2019-04-30","paper":"/paper/mixhop-higher-order-graph-convolution","paper_url":"https://arxiv.org/abs/1905.00067v3","paper_title":"MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/mixhop","n_code_links":3,"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":33,"rows_with_any_sample_ran":28,"distinct_papers_with_graph_line":13,"distinct_papers_with_any_sample_ran":10,"samples_over_distinct_papers":{"n_ran":121,"n_unverified":123,"n_samples":244,"n_pointer_only_licence":90,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":190,"n_unverified":185,"n_samples":375,"n_pointer_only_licence":94,"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"}}}