{"url":"/sota/hate-speech-detection-on-ethos-binary","task":{"name":"Hate Speech Detection","url":"/task/hate-speech-detection","note":null},"dataset":{"name":"Ethos Binary","url":"/dataset/ethos"},"category":"Natural Language Processing","categories":["Natural Language Processing"],"category_note":null,"description":"Hate speech detection is the task of detecting if communication such as text, audio, and so on contains hatred and or encourages violence towards a person or a group of people. This is usually based on prejudice against 'protected characteristics' such as their ethnicity, gender, sexual orientation, religion, age et al. Some example benchmarks are ETHOS and HateXplain. Models can be evaluated with metrics like the F-score or F-measure.","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":["F1-score","Classification Accuracy","Precision"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"F1-score":"higher","Classification Accuracy":"higher","Precision":"higher"}},"counts":{"rows":12,"rows_with_code":11,"rows_with_paper_page":12,"rows_dated":12,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"BiLSTM + static BE","metrics":{"Classification Accuracy":"0.8015","F1-score":"0.7971","Precision":"0.8037"},"uses_additional_data":false,"paper_date":"2021-06-29","paper":"/paper/hate-speech-detection-using-static-bert","paper_url":"https://arxiv.org/abs/2106.15537v1","paper_title":"Hate speech detection using static BERT embeddings","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"BERT","metrics":{"Classification Accuracy":"0.7664","F1-score":"0.7883","Precision":"79.17"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/ethos-an-online-hate-speech-detection-dataset","paper_url":"https://arxiv.org/abs/2006.08328v2","paper_title":"ETHOS: an Online Hate Speech Detection Dataset","code":"https://github.com/intelligence-csd-auth-gr/Ethos-Hate-Speech-Dataset","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"BiLSTM+Attention+FT","metrics":{"Classification Accuracy":"0.7734","F1-score":"0.768","Precision":"77.76"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/ethos-an-online-hate-speech-detection-dataset","paper_url":"https://arxiv.org/abs/2006.08328v2","paper_title":"ETHOS: an Online Hate Speech Detection Dataset","code":"https://github.com/intelligence-csd-auth-gr/Ethos-Hate-Speech-Dataset","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"OPT-175B (few-shot)","metrics":{"F1-score":"0.759"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}},{"rank_in_archive_order":5,"model":"CNN+Attention+FT+GV","metrics":{"Classification Accuracy":"0.7515","F1-score":"0.7441","Precision":"74.92"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/ethos-an-online-hate-speech-detection-dataset","paper_url":"https://arxiv.org/abs/2006.08328v2","paper_title":"ETHOS: an Online Hate Speech Detection Dataset","code":"https://github.com/intelligence-csd-auth-gr/Ethos-Hate-Speech-Dataset","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"OPT-175B (one-shot)","metrics":{"F1-score":"0.713"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}},{"rank_in_archive_order":7,"model":"OPT-175B (zero-shot)","metrics":{"F1-score":"0.667"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}},{"rank_in_archive_order":8,"model":"SVM","metrics":{"Classification Accuracy":"0.6643","F1-score":"0.6607","Precision":"66.47"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/ethos-an-online-hate-speech-detection-dataset","paper_url":"https://arxiv.org/abs/2006.08328v2","paper_title":"ETHOS: an Online Hate Speech Detection Dataset","code":"https://github.com/intelligence-csd-auth-gr/Ethos-Hate-Speech-Dataset","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"Random Forests","metrics":{"Classification Accuracy":"0.6504","F1-score":"0.6441","Precision":"64.69"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/ethos-an-online-hate-speech-detection-dataset","paper_url":"https://arxiv.org/abs/2006.08328v2","paper_title":"ETHOS: an Online Hate Speech Detection Dataset","code":"https://github.com/intelligence-csd-auth-gr/Ethos-Hate-Speech-Dataset","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"Davinci (zero-shot)","metrics":{"F1-score":"0.628"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}},{"rank_in_archive_order":11,"model":"Davinci (one-shot)","metrics":{"F1-score":"0.616"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}},{"rank_in_archive_order":12,"model":"Davinci (few-shot)","metrics":{"F1-score":"0.354"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/opt-open-pre-trained-transformer-language","paper_url":"https://arxiv.org/abs/2205.01068v4","paper_title":"OPT: Open Pre-trained Transformer Language Models","code":"https://github.com/facebookresearch/metaseq","n_code_links":11,"syntology":{"n_ran":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"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":6,"n_unverified":18,"n_samples":24,"n_pointer_only_licence":16,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":36,"n_unverified":108,"n_samples":144,"n_pointer_only_licence":96,"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"}}}