{"url":"/sota/single-step-retrosynthesis-on-uspto-50k","task":{"name":"Single-step retrosynthesis","url":"/task/single-step-retrosynthesis","note":null},"dataset":{"name":"USPTO-50k","url":"/dataset/uspto-50k"},"category":"Medical","categories":["Medical"],"category_note":null,"description":null,"description_from":null,"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":["Top-1 accuracy","Top-3 accuracy","Top-5 accuracy","Top-10 accuracy","Top-20 accuracy","Top-50 accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Top-1 accuracy":"higher","Top-3 accuracy":"higher","Top-5 accuracy":"higher","Top-10 accuracy":"higher","Top-20 accuracy":"higher","Top-50 accuracy":"higher"}},"counts":{"rows":35,"rows_with_code":23,"rows_with_paper_page":35,"rows_dated":35,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"NAG2G (reaction class as prior)","metrics":{"Top-1 accuracy":"67.2","Top-10 accuracy":"93.8","Top-3 accuracy":"86.4","Top-5 accuracy":"90.5"},"uses_additional_data":false,"paper_date":"2023-09-27","paper":"/paper/node-aligned-graph-to-graph-generation-for","paper_url":"https://arxiv.org/abs/2309.15798v2","paper_title":"Node-Aligned Graph-to-Graph (NAG2G): Elevating Template-Free Deep Learning Approaches in Single-Step Retrosynthesis","code":"https://github.com/dptech-corp/nag2g","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":2,"model":"UAlign (reaction class as prior)","metrics":{"Top-1 accuracy":"66.4","Top-10 accuracy":"95.0","Top-3 accuracy":"86.7","Top-5 accuracy":"91.5"},"uses_additional_data":false,"paper_date":"2024-03-25","paper":"/paper/ualign-pushing-the-limit-of-template-free","paper_url":"https://arxiv.org/abs/2404.00044v2","paper_title":"UAlign: Pushing the Limit of Template-free Retrosynthesis Prediction with Unsupervised SMILES Alignment","code":"https://github.com/zengkaipeng/UAlign","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"MARS (reaction class as prior)","metrics":{"Top-1 accuracy":"66.2","Top-10 accuracy":"92.9","Top-3 accuracy":"85.8","Top-5 accuracy":"90.2"},"uses_additional_data":false,"paper_date":"2022-09-27","paper":"/paper/mars-a-motif-based-autoregressive-model-for","paper_url":"https://arxiv.org/abs/2209.13178v1","paper_title":"MARS: A Motif-based Autoregressive Model for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"RetroComposer (reaction class as prior)","metrics":{"Top-1 accuracy":"65.9","Top-10 accuracy":"91.5","Top-3 accuracy":"85.8","Top-5 accuracy":"89.5"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/retrocomposer-discovering-novel-reactions-by","paper_url":"https://arxiv.org/abs/2112.11225v2","paper_title":"RetroComposer: Composing Templates for Template-Based Retrosynthesis Prediction","code":"https://github.com/uta-smile/retrocomposer","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"SemiRetro (reaction class as prior)","metrics":{"Top-1 accuracy":"65.8","Top-10 accuracy":"92.8","Top-3 accuracy":"85.7","Top-5 accuracy":"89.8"},"uses_additional_data":false,"paper_date":"2022-02-12","paper":"/paper/semiretro-semi-template-framework-boosts-deep-1","paper_url":"https://arxiv.org/abs/2202.08205v1","paper_title":"SemiRetro: Semi-template framework boosts deep retrosynthesis prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"O-GNN (reaction class as prior)","metrics":{"Top-1 accuracy":"65.7","Top-10 accuracy":"96.9","Top-3 accuracy":"87.7","Top-5 accuracy":"93.4","Top-50 accuracy":"98.3"},"uses_additional_data":false,"paper_date":"2023-05-01","paper":"/paper/o-gnn-incorporating-ring-priors-into","paper_url":"https://openreview.net/forum?id=5cFfz6yMVPU","paper_title":"O-GNN: Incorporating Ring Priors into Molecular Modeling","code":"https://github.com/O-GNN/O-GNN","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"Dual-TF (reaction class as prior)","metrics":{"Top-1 accuracy":"65.7","Top-10 accuracy":"85.9","Top-3 accuracy":"81.9","Top-5 accuracy":"84.7"},"uses_additional_data":false,"paper_date":"2020-07-14","paper":"/paper/energy-based-view-of-retrosynthesis","paper_url":"https://arxiv.org/abs/2007.13437v2","paper_title":"Energy-based View of Retrosynthesis","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"LocalRetro (reaction class as prior)","metrics":{"Top-1 accuracy":"63.9","Top-10 accuracy":"96.3","Top-3 accuracy":"86.1","Top-5 accuracy":"92.4","Top-50 accuracy":"97.9"},"uses_additional_data":false,"paper_date":"2021-08-05","paper":"/paper/deep-retrosynthetic-reaction-prediction-using","paper_url":"https://pubs.acs.org/doi/10.1021/jacsau.1c00246","paper_title":"Deep Retrosynthetic Reaction Prediction using Local Reactivity and Global Attention","code":"https://github.com/kaist-amsg/LocalRetro","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"GraphRetro (reaction class as prior)","metrics":{"Top-1 accuracy":"63.9","Top-10 accuracy":"88.1","Top-3 accuracy":"81.5","Top-5 accuracy":"85.2"},"uses_additional_data":false,"paper_date":"2021-06-04","paper":"/paper/learning-graph-models-for-template-free-1","paper_url":"https://arxiv.org/abs/2006.07038","paper_title":"Learning Graph Models for Template-Free Retrosynthesis","code":null,"n_code_links":0,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":1}},{"rank_in_archive_order":10,"model":"MEGAN (reaction class as prior)","metrics":{"Top-1 accuracy":"60.7","Top-10 accuracy":"91.6","Top-3 accuracy":"82","Top-5 accuracy":"87.5"},"uses_additional_data":false,"paper_date":"2020-06-27","paper":"/paper/molecule-edit-graph-attention-network","paper_url":"https://arxiv.org/abs/2006.15426v2","paper_title":"Molecule Edit Graph Attention Network: Modeling Chemical Reactions as Sequences of Graph Edits","code":"https://github.com/molecule-one/megan","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"GET-LT1 (reaction class as prior)","metrics":{"Top-1 accuracy":"57.4","Top-10 accuracy":"77.4","Top-3 accuracy":"71.3","Top-5 accuracy":"74.8"},"uses_additional_data":false,"paper_date":"2019-09-25","paper":"/paper/molecular-graph-enhanced-transformer-for","paper_url":"https://openreview.net/forum?id=S1e__ANKvB","paper_title":"Molecular Graph Enhanced Transformer for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"GET-LT2 (reaction class as prior)","metrics":{"Top-1 accuracy":"56.2","Top-10 accuracy":"74.7","Top-3 accuracy":"69.4","Top-5 accuracy":"72.5"},"uses_additional_data":false,"paper_date":"2019-09-25","paper":"/paper/molecular-graph-enhanced-transformer-for","paper_url":"https://openreview.net/forum?id=S1e__ANKvB","paper_title":"Molecular Graph Enhanced Transformer for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"NAG2G (reaction class unknown)","metrics":{"Top-1 accuracy":"55.1","Top-10 accuracy":"89.9","Top-3 accuracy":"76.9","Top-5 accuracy":"83.4"},"uses_additional_data":false,"paper_date":"2023-09-27","paper":"/paper/node-aligned-graph-to-graph-generation-for","paper_url":"https://arxiv.org/abs/2309.15798v2","paper_title":"Node-Aligned Graph-to-Graph (NAG2G): Elevating Template-Free Deep Learning Approaches in Single-Step Retrosynthesis","code":"https://github.com/dptech-corp/nag2g","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":14,"model":"SemiRetro (reaction class unknown)","metrics":{"Top-1 accuracy":"54.9","Top-10 accuracy":"84.1","Top-3 accuracy":"75.3","Top-5 accuracy":"80.4"},"uses_additional_data":false,"paper_date":"2022-02-12","paper":"/paper/semiretro-semi-template-framework-boosts-deep-1","paper_url":"https://arxiv.org/abs/2202.08205v1","paper_title":"SemiRetro: Semi-template framework boosts deep retrosynthesis prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"MARS (reaction class unknown)","metrics":{"Top-1 accuracy":"54.6","Top-10 accuracy":"88.5","Top-3 accuracy":"76.4","Top-5 accuracy":"83.3"},"uses_additional_data":false,"paper_date":"2022-09-27","paper":"/paper/mars-a-motif-based-autoregressive-model-for","paper_url":"https://arxiv.org/abs/2209.13178v1","paper_title":"MARS: A Motif-based Autoregressive Model for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"RetroComposer (reaction class unknown)","metrics":{"Top-1 accuracy":"54.4","Top-10 accuracy":"87.7","Top-3 accuracy":"77.2","Top-5 accuracy":"83.2"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/retrocomposer-discovering-novel-reactions-by","paper_url":"https://arxiv.org/abs/2112.11225v2","paper_title":"RetroComposer: Composing Templates for Template-Based Retrosynthesis Prediction","code":"https://github.com/uta-smile/retrocomposer","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"Chemformer-Large (reaction class unknown)","metrics":{"Top-1 accuracy":"54.3","Top-10 accuracy":"63.0","Top-5 accuracy":"62.3"},"uses_additional_data":false,"paper_date":"2022-01-31","paper":"/paper/chemformer-a-pre-trained-transformer-for","paper_url":"https://iopscience.iop.org/article/10.1088/2632-2153/ac3ffb","paper_title":"Chemformer: a pre-trained transformer for computational chemistry","code":"https://github.com/MolecularAI/Chemformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"O-GNN (reaction class unknown)","metrics":{"Top-1 accuracy":"54.1","Top-10 accuracy":"92.5","Top-3 accuracy":"77.7","Top-5 accuracy":"86.0","Top-50 accuracy":"98.2"},"uses_additional_data":false,"paper_date":"2023-05-01","paper":"/paper/o-gnn-incorporating-ring-priors-into","paper_url":"https://openreview.net/forum?id=5cFfz6yMVPU","paper_title":"O-GNN: Incorporating Ring Priors into Molecular Modeling","code":"https://github.com/O-GNN/O-GNN","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"G2GT (reaction class unknown)","metrics":{"Top-1 accuracy":"54.1","Top-10 accuracy":"77.7","Top-3 accuracy":"69.9","Top-5 accuracy":"74.5"},"uses_additional_data":false,"paper_date":"2022-04-19","paper":"/paper/g2gt-retrosynthesis-prediction-with-graph-to","paper_url":"https://arxiv.org/abs/2204.08608v1","paper_title":"G2GT: Retrosynthesis Prediction with Graph to Graph Attention Neural Network and Self-Training","code":"https://github.com/Anonnoname/G2GT_2","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"GraphRetro (reaction class unknown)","metrics":{"Top-1 accuracy":"53.7","Top-10 accuracy":"75.5","Top-3 accuracy":"68.3","Top-5 accuracy":"72.2"},"uses_additional_data":false,"paper_date":"2021-06-04","paper":"/paper/learning-graph-models-for-template-free-1","paper_url":"https://arxiv.org/abs/2006.07038","paper_title":"Learning Graph Models for Template-Free Retrosynthesis","code":null,"n_code_links":0,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":1}},{"rank_in_archive_order":21,"model":"Dual-TF (reaction class unknown)","metrics":{"Top-1 accuracy":"53.6","Top-10 accuracy":"77.0","Top-3 accuracy":"70.7","Top-5 accuracy":"74.6"},"uses_additional_data":false,"paper_date":"2020-07-14","paper":"/paper/energy-based-view-of-retrosynthesis","paper_url":"https://arxiv.org/abs/2007.13437v2","paper_title":"Energy-based View of Retrosynthesis","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"UAlign (reaction class unknown)","metrics":{"Top-1 accuracy":"53.5","Top-10 accuracy":"90.5","Top-3 accuracy":"77.3","Top-5 accuracy":"84.6"},"uses_additional_data":false,"paper_date":"2024-03-25","paper":"/paper/ualign-pushing-the-limit-of-template-free","paper_url":"https://arxiv.org/abs/2404.00044v2","paper_title":"UAlign: Pushing the Limit of Template-free Retrosynthesis Prediction with Unsupervised SMILES Alignment","code":"https://github.com/zengkaipeng/UAlign","n_code_links":1,"syntology":null},{"rank_in_archive_order":23,"model":"Augmented Transformer (reaction class unknown)","metrics":{"Top-1 accuracy":"53.5","Top-10 accuracy":"85.7","Top-5 accuracy":"81.0"},"uses_additional_data":false,"paper_date":"2020-03-05","paper":"/paper/augmented-transformer-achieves-97-and-85-for","paper_url":"https://arxiv.org/abs/2003.02804v2","paper_title":"State-of-the-Art Augmented NLP Transformer models for direct and single-step retrosynthesis","code":"https://github.com/bigchem/synthesis","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"LocalRetro (reaction class unknown)","metrics":{"Top-1 accuracy":"53.4","Top-10 accuracy":"92.4","Top-3 accuracy":"77.5","Top-5 accuracy":"85.9","Top-50 accuracy":"97.7"},"uses_additional_data":false,"paper_date":"2021-08-05","paper":"/paper/deep-retrosynthetic-reaction-prediction-using","paper_url":"https://pubs.acs.org/doi/10.1021/jacsau.1c00246","paper_title":"Deep Retrosynthetic Reaction Prediction using Local Reactivity and Global Attention","code":"https://github.com/kaist-amsg/LocalRetro","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"RetroSim (reaction class as prior)","metrics":{"Top-1 accuracy":"52.9","Top-10 accuracy":"88.1","Top-3 accuracy":"73.8","Top-5 accuracy":"81.2"},"uses_additional_data":false,"paper_date":"2017-11-16","paper":"/paper/computer-assisted-retrosynthesis-based-on","paper_url":"https://pubs.acs.org/doi/10.1021/acscentsci.7b00355","paper_title":"Computer-Assisted Retrosynthesis Based on Molecular Similarity","code":"https://github.com/connorcoley/retrosim","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"Graph2SMILES-D-GCN (reaction class unknown)","metrics":{"Top-1 accuracy":"52.9","Top-10 accuracy":"72.9","Top-3 accuracy":"66.5","Top-5 accuracy":"70.0"},"uses_additional_data":false,"paper_date":"2021-10-19","paper":"/paper/permutation-invariant-graph-to-sequence-model-1","paper_url":"https://arxiv.org/abs/2110.09681v1","paper_title":"Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction","code":"https://github.com/coleygroup/graph2smiles","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"GLN","metrics":{"Top-1 accuracy":"52.5","Top-10 accuracy":"83.7","Top-20 accuracy":"89","Top-3 accuracy":"69","Top-5 accuracy":"75.6","Top-50 accuracy":"92.4"},"uses_additional_data":false,"paper_date":"2020-01-06","paper":"/paper/retrosynthesis-prediction-with-conditional-1","paper_url":"https://arxiv.org/abs/2001.01408v1","paper_title":"Retrosynthesis Prediction with Conditional Graph Logic Network","code":"https://github.com/Hanjun-Dai/GLN","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":28,"model":"MHNreact (reaction class unknown)","metrics":{"Top-1 accuracy":"51.8","Top-10 accuracy":"88.1","Top-20 accuracy":"92.0","Top-3 accuracy":"74.6","Top-5 accuracy":"81.2","Top-50 accuracy":"94.0"},"uses_additional_data":false,"paper_date":"2022-01-15","paper":"/paper/improving-few-and-zero-shot-reaction-template","paper_url":"https://pubs.acs.org/doi/abs/10.1021/acs.jcim.1c01065","paper_title":"Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks","code":"https://github.com/ml-jku/mhn-react","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"RetroPrime (reaction class unknown)","metrics":{"Top-1 accuracy":"51.4","Top-10 accuracy":"76.1","Top-3 accuracy":"70.8","Top-5 accuracy":"74.0"},"uses_additional_data":false,"paper_date":"2021-09-15","paper":"/paper/retroprime-a-diverse-plausible-and","paper_url":"https://www.sciencedirect.com/science/article/pii/S1385894721014303?via%3Dihub","paper_title":"RetroPrime: A Diverse, plausible and Transformer-based method for Single-Step retrosynthesis predictions","code":"https://github.com/wangxr0526/RetroPrime","n_code_links":1,"syntology":null},{"rank_in_archive_order":30,"model":"Graph2SMILES-D-GAT (reaction class unknown)","metrics":{"Top-1 accuracy":"51.2","Top-10 accuracy":"73.9","Top-3 accuracy":"66.3","Top-5 accuracy":"70.4"},"uses_additional_data":false,"paper_date":"2021-10-19","paper":"/paper/permutation-invariant-graph-to-sequence-model-1","paper_url":"https://arxiv.org/abs/2110.09681v1","paper_title":"Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction","code":"https://github.com/coleygroup/graph2smiles","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":31,"model":"G2Gs","metrics":{"Top-1 accuracy":"48.9","Top-10 accuracy":"75.5","Top-3 accuracy":"67.6","Top-5 accuracy":"72.5"},"uses_additional_data":false,"paper_date":"2020-03-28","paper":"/paper/a-graph-to-graphs-framework-for","paper_url":"https://arxiv.org/abs/2003.12725v3","paper_title":"A Graph to Graphs Framework for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"MEGAN (reaction class unknown)","metrics":{"Top-1 accuracy":"48.1","Top-10 accuracy":"86.1","Top-20 accuracy":"90.3","Top-3 accuracy":"70.7","Top-5 accuracy":"78.4","Top-50 accuracy":"93.2"},"uses_additional_data":false,"paper_date":"2020-06-27","paper":"/paper/molecule-edit-graph-attention-network","paper_url":"https://arxiv.org/abs/2006.15426v2","paper_title":"Molecule Edit Graph Attention Network: Modeling Chemical Reactions as Sequences of Graph Edits","code":"https://github.com/molecule-one/megan","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":33,"model":"GET-LT1 (reaction class unknown)","metrics":{"Top-1 accuracy":"44.9","Top-10 accuracy":"65.9","Top-3 accuracy":"58.8","Top-5 accuracy":"62.4"},"uses_additional_data":false,"paper_date":"2019-09-25","paper":"/paper/molecular-graph-enhanced-transformer-for","paper_url":"https://openreview.net/forum?id=S1e__ANKvB","paper_title":"Molecular Graph Enhanced Transformer for Retrosynthesis Prediction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":34,"model":"RetroSim (reaction class unknown)","metrics":{"Top-1 accuracy":"37.3","Top-10 accuracy":"74.1","Top-3 accuracy":"54.7","Top-5 accuracy":"63.3"},"uses_additional_data":false,"paper_date":"2017-11-16","paper":"/paper/computer-assisted-retrosynthesis-based-on","paper_url":"https://pubs.acs.org/doi/10.1021/acscentsci.7b00355","paper_title":"Computer-Assisted Retrosynthesis Based on Molecular Similarity","code":"https://github.com/connorcoley/retrosim","n_code_links":1,"syntology":null},{"rank_in_archive_order":35,"model":"Expert System (reaction class as prior)","metrics":{"Top-1 accuracy":"35.2","Top-10 accuracy":"65.1","Top-3 accuracy":"52.3","Top-5 accuracy":"59.1"},"uses_additional_data":false,"paper_date":"2017-06-06","paper":"/paper/retrosynthetic-reaction-prediction-using","paper_url":"http://arxiv.org/abs/1706.01643v1","paper_title":"Retrosynthetic reaction prediction using neural sequence-to-sequence models","code":"https://github.com/pandegroup/reaction_prediction_seq2seq","n_code_links":1,"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":9,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":10,"n_unverified":14,"n_samples":24,"n_pointer_only_licence":4,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":19,"n_unverified":25,"n_samples":44,"n_pointer_only_licence":8,"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"}}}