{"url":"/sota/toxic-comment-classification-on-civil","task":{"name":"Toxic Comment Classification","url":"/task/toxic-comment-classification","note":null},"dataset":{"name":"Civil Comments","url":"/dataset/civil-comments"},"category":"Natural Language Processing","categories":["Natural Language Processing"],"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":["GMB BPSN","GMB Subgroup","Micro F1","Precision","Recall","Macro F1","AUROC","GMB BNSP"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"GMB BPSN":null,"GMB Subgroup":null,"Micro F1":"higher","Precision":"higher","Recall":"higher","Macro F1":"higher","AUROC":null,"GMB BNSP":null}},"counts":{"rows":22,"rows_with_code":22,"rows_with_paper_page":22,"rows_dated":22,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"RoBERTa Focal Loss","metrics":{"AUROC":"0.9818","GMB BNSP":"0.9581","GMB BPSN":"0.901","GMB Subgroup":"0.8807","Macro F1":"0.4648","Micro F1":"0.5524","Precision":"0.4017","Recall":"0.8839"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"AlBERT","metrics":{"AUROC":"0.979","GMB BNSP":"0.9499","GMB BPSN":"0.8982","GMB Subgroup":"0.8734","Macro F1":"0.3541","Micro F1":"0.4845","Precision":"0.3247","Recall":"0.9104"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"BERTweet","metrics":{"AUROC":"0.979","GMB BNSP":"0.9603","GMB BPSN":"0.8945","GMB Subgroup":"0.878","Macro F1":"0.3612","Micro F1":"0.4928","Precision":"0.3363","Recall":"0.9216"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"HateBERT","metrics":{"AUROC":"0.9791","GMB BNSP":"0.9589","GMB BPSN":"0.8915","GMB Subgroup":"0.8744","Macro F1":"0.3679","Micro F1":"0.4844","Precision":"0.3297","Recall":"0.9165"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"RoBERTa BCE","metrics":{"AUROC":"0.9813","GMB BNSP":"0.9616","GMB BPSN":"0.8901","GMB Subgroup":"0.88","Macro F1":"0.4749","Micro F1":"0.5359","Precision":"0.3836","Recall":"0.8891"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"XLM RoBERTa","metrics":{"GMB BPSN":"0.8859","Micro F1":"0.468","Precision":"0.3135","Recall":"0.923"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"XLNet","metrics":{"GMB BNSP":"0.9597","GMB BPSN":"0.8834","GMB Subgroup":"0.8689","Macro F1":"0.3336","Micro F1":"0.4586","Precision":"0.3045","Recall":"0.9254"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"DistilBERT","metrics":{"AUROC":"0.9804","GMB BNSP":"0.9644","GMB BPSN":"0.874","GMB Subgroup":"0.8762","Macro F1":"0.3879","Micro F1":"0.5115","Precision":"0.3572","Recall":"0.9001"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"BiGRU","metrics":{"GMB BPSN":"0.8616"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"Unfreeze Glove ResNet 44","metrics":{"AUROC":"0.966","GMB BPSN":"0.8493","GMB Subgroup":"0.8421","Macro F1":"0.4648","Micro F1":"0.5958","Precision":"0.4835","Recall":"0.7759"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"Unfreeze Glove ResNet 56","metrics":{"AUROC":"0.9639","GMB BPSN":"0.8445","GMB Subgroup":"0.8487","Macro F1":"0.3778","Recall":"0.8707"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"Compact Convolutional Transformer (CCT)","metrics":{"AUROC":"0.9526","GMB BNSP":"0.9447","GMB BPSN":"0.8307","GMB Subgroup":"0.8133","Macro F1":"0.3428","Micro F1":"0.4874","Precision":"0.3507","Recall":"0.7983"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"Freeze Glove ResNet 44","metrics":{"GMB BPSN":"0.7876","GMB Subgroup":"0.8219","Macro F1":"0.4189","Micro F1":"0.5591","Precision":"0.4631","Recall":"0.7053"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"BiLSTM","metrics":{"GMB Subgroup":"0.8636","Micro F1":"0.5115","Precision":"0.3572"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/a-benchmark-for-toxic-comment-classification","paper_url":"https://arxiv.org/abs/2301.11125v1","paper_title":"A benchmark for toxic comment classification on Civil Comments dataset","code":"https://github.com/Nigiva/hatespeech-detection-models","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"ResNet + RoBERTa finetune","metrics":{"AUROC":"0.97"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"Trompt + OpenAI embedding","metrics":{"AUROC":"0.947"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"ResNet + OpenAI embedding","metrics":{"AUROC":"0.945"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"Trompt + RoBERTa embedding","metrics":{"AUROC":"0.885"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"ResNet + RoBERTa embedding","metrics":{"AUROC":"0.882"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"LightGBM + RoBERTa embedding","metrics":{"AUROC":"0.865"},"uses_additional_data":false,"paper_date":"2024-03-31","paper":"/paper/pytorch-frame-a-modular-framework-for-multi","paper_url":"https://arxiv.org/abs/2404.00776v2","paper_title":"PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning","code":"https://github.com/pyg-team/pytorch-frame","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"PaLM 2 (few-shot, k=10)","metrics":{"AUROC":"0.8535"},"uses_additional_data":false,"paper_date":"2023-05-17","paper":"/paper/palm-2-technical-report-1","paper_url":"https://arxiv.org/abs/2305.10403v3","paper_title":"PaLM 2 Technical Report","code":"https://github.com/eternityyw/tram-benchmark","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"PaLM 2 (zero-shot)","metrics":{"AUROC":"0.7596"},"uses_additional_data":false,"paper_date":"2023-05-17","paper":"/paper/palm-2-technical-report-1","paper_url":"https://arxiv.org/abs/2305.10403v3","paper_title":"PaLM 2 Technical Report","code":"https://github.com/eternityyw/tram-benchmark","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 6,264 of the 9,581 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":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+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":6,"rows_with_any_sample_ran":6,"distinct_papers_with_graph_line":1,"distinct_papers_with_any_sample_ran":1,"samples_over_distinct_papers":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":12,"n_unverified":0,"n_samples":12,"n_pointer_only_licence":0,"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"}}}