{"url":"/sota/chinese-named-entity-recognition-on-msra","task":{"name":"Chinese Named Entity Recognition","url":"/task/chinese-named-entity-recognition","note":null},"dataset":{"name":"MSRA","url":"/dataset/msra-cn-ner"},"category":"Natural Language Processing","categories":["Natural Language Processing"],"category_note":null,"description":"Chinese named entity recognition is a subtask of information extraction that seeks to locate and classify named entities mentioned in unstructured text into pre-defined categories such as person names, organizations, locations, medical codes, time expressions, quantities, monetary values, percentages, etc. from Chinese text (Source: Adapted from Wikipedia).","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","Precision","Recall"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"F1":"higher","Precision":"higher","Recall":"higher"}},"counts":{"rows":21,"rows_with_code":21,"rows_with_paper_page":21,"rows_dated":21,"rows_using_additional_data":3},"rows":[{"rank_in_archive_order":1,"model":"BERT-MRC+DSC","metrics":{"F1":"96.72"},"uses_additional_data":true,"paper_date":"2019-11-07","paper":"/paper/dice-loss-for-data-imbalanced-nlp-tasks","paper_url":"https://arxiv.org/abs/1911.02855v3","paper_title":"Dice Loss for Data-imbalanced NLP Tasks","code":"https://github.com/ShannonAI/dice_loss_for_NLP","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"BERT-CRF (Replicated in AdaSeq)","metrics":{"F1":"96.69"},"uses_additional_data":true,"paper_date":"2021-05-08","paper":"/paper/improving-named-entity-recognition-by","paper_url":"https://arxiv.org/abs/2105.03654v3","paper_title":"Improving Named Entity Recognition by External Context Retrieving and Cooperative Learning","code":"https://github.com/modelscope/adaseq","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":"Baseline + BS","metrics":{"F1":"96.26"},"uses_additional_data":false,"paper_date":"2022-04-26","paper":"/paper/boundary-smoothing-for-named-entity-1","paper_url":"https://arxiv.org/abs/2204.12031v1","paper_title":"Boundary Smoothing for Named Entity Recognition","code":"https://github.com/syuoni/eznlp","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"W2NER","metrics":{"F1":"96.10"},"uses_additional_data":false,"paper_date":"2021-12-19","paper":"/paper/unified-named-entity-recognition-as-word-word","paper_url":"https://arxiv.org/abs/2112.10070v1","paper_title":"Unified Named Entity Recognition as Word-Word Relation Classification","code":"https://github.com/ljynlp/w2ner","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"FLAT+BERT","metrics":{"F1":"96.09"},"uses_additional_data":false,"paper_date":"2020-04-24","paper":"/paper/flat-chinese-ner-using-flat-lattice","paper_url":"https://arxiv.org/abs/2004.11795v2","paper_title":"FLAT: Chinese NER Using Flat-Lattice Transformer","code":"https://github.com/LeeSureman/Flat-Lattice-Transformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"BERT-MRC","metrics":{"F1":"95.75"},"uses_additional_data":true,"paper_date":"2019-10-25","paper":"/paper/a-unified-mrc-framework-for-named-entity","paper_url":"https://arxiv.org/abs/1910.11476v7","paper_title":"A Unified MRC Framework for Named Entity Recognition","code":"https://github.com/ShannonAI/mrc-for-flat-nested-ner","n_code_links":8,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":2}},{"rank_in_archive_order":7,"model":"FGN","metrics":{"F1":"95.64"},"uses_additional_data":false,"paper_date":"2020-01-15","paper":"/paper/fgn-fusion-glyph-network-for-chinese-named","paper_url":"https://arxiv.org/abs/2001.05272v6","paper_title":"FGN: Fusion Glyph Network for Chinese Named Entity Recognition","code":"https://github.com/AidenHuen/FGN-NER","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"Glyce + BERT","metrics":{"F1":"95.54","Precision":"95.57","Recall":"95.51"},"uses_additional_data":false,"paper_date":"2019-01-29","paper":"/paper/glyce-glyph-vectors-for-chinese-character","paper_url":"https://arxiv.org/abs/1901.10125v5","paper_title":"Glyce: Glyph-vectors for Chinese Character Representations","code":"https://github.com/ShannonAI/glyce","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"ZEN (Init with Chinese BERT)","metrics":{"F1":"95.25"},"uses_additional_data":false,"paper_date":"2019-11-02","paper":"/paper/zen-pre-training-chinese-text-encoder","paper_url":"https://arxiv.org/abs/1911.00720v1","paper_title":"ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations","code":"https://github.com/sinovation/ZEN","n_code_links":7,"syntology":{"n_ran":7,"n_unverified":28,"n_samples":35,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"ERNIE 2.0 Large","metrics":{"F1":"95"},"uses_additional_data":false,"paper_date":"2019-07-29","paper":"/paper/ernie-20-a-continual-pre-training-framework","paper_url":"https://arxiv.org/abs/1907.12412v2","paper_title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","code":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":11,"model":"DiffusionNER","metrics":{"F1":"94.91"},"uses_additional_data":false,"paper_date":"2023-05-22","paper":"/paper/diffusionner-boundary-diffusion-for-named","paper_url":"https://arxiv.org/abs/2305.13298v1","paper_title":"DiffusionNER: Boundary Diffusion for Named Entity Recognition","code":"https://github.com/tricktreat/diffusionner","n_code_links":3,"syntology":null},{"rank_in_archive_order":12,"model":"NFLAT","metrics":{"F1":"94.55","Precision":"94.92","Recall":"94.19"},"uses_additional_data":false,"paper_date":"2022-05-12","paper":"/paper/nflat-non-flat-lattice-transformer-for","paper_url":"https://arxiv.org/abs/2205.05832v3","paper_title":"NFLAT: Non-Flat-Lattice Transformer for Chinese Named Entity Recognition","code":"https://github.com/codermusou/nflat4cner","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"FLAT","metrics":{"F1":"94.12"},"uses_additional_data":false,"paper_date":"2020-04-24","paper":"/paper/flat-chinese-ner-using-flat-lattice","paper_url":"https://arxiv.org/abs/2004.11795v2","paper_title":"FLAT: Chinese NER Using Flat-Lattice Transformer","code":"https://github.com/LeeSureman/Flat-Lattice-Transformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"ERNIE","metrics":{"F1":"93.8"},"uses_additional_data":false,"paper_date":"2019-04-19","paper":"/paper/ernie-enhanced-representation-through","paper_url":"http://arxiv.org/abs/1904.09223v1","paper_title":"ERNIE: Enhanced Representation through Knowledge Integration","code":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0","n_code_links":19,"syntology":{"n_ran":0,"n_unverified":7,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"ERNIE 2.0 Base","metrics":{"F1":"93.8"},"uses_additional_data":false,"paper_date":"2019-07-29","paper":"/paper/ernie-20-a-continual-pre-training-framework","paper_url":"https://arxiv.org/abs/1907.12412v2","paper_title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","code":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":16,"model":"LSTM + Lexicon augment","metrics":{"F1":"93.5"},"uses_additional_data":false,"paper_date":"2019-08-16","paper":"/paper/simplify-the-usage-of-lexicon-in-chinese-ner","paper_url":"https://arxiv.org/abs/1908.05969v2","paper_title":"Simplify the Usage of Lexicon in Chinese NER","code":"https://github.com/v-mipeng/LexiconAugmentedNER","n_code_links":2,"syntology":null},{"rank_in_archive_order":17,"model":"PIQN","metrics":{"F1":"93.48"},"uses_additional_data":false,"paper_date":"2022-03-20","paper":"/paper/parallel-instance-query-network-for-named","paper_url":"https://arxiv.org/abs/2203.10545v1","paper_title":"Parallel Instance Query Network for Named Entity Recognition","code":"https://github.com/tricktreat/piqn","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"ZEN (Random Init)","metrics":{"F1":"93.24"},"uses_additional_data":false,"paper_date":"2019-11-02","paper":"/paper/zen-pre-training-chinese-text-encoder","paper_url":"https://arxiv.org/abs/1911.00720v1","paper_title":"ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations","code":"https://github.com/sinovation/ZEN","n_code_links":7,"syntology":{"n_ran":7,"n_unverified":28,"n_samples":35,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"Lattice","metrics":{"F1":"93.18"},"uses_additional_data":false,"paper_date":"2018-05-05","paper":"/paper/chinese-ner-using-lattice-lstm","paper_url":"http://arxiv.org/abs/1805.02023v4","paper_title":"Chinese NER Using Lattice LSTM","code":"https://github.com/jiesutd/LatticeLSTM","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":20,"model":"CAN-NER Model","metrics":{"F1":"92.97","Precision":"93.53","Recall":"92.42"},"uses_additional_data":false,"paper_date":"2019-04-03","paper":"/paper/can-ner-convolutional-attention-network","paper_url":"https://arxiv.org/abs/1904.02141v3","paper_title":"CAN-NER: Convolutional Attention Network for Chinese Named Entity Recognition","code":"https://github.com/microsoft/vert-papers/tree/master/papers/CAN-NER","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"TENER","metrics":{"F1":"92.74"},"uses_additional_data":false,"paper_date":"2019-11-10","paper":"/paper/tener-adapting-transformer-encoder-for-name","paper_url":"https://arxiv.org/abs/1911.04474v3","paper_title":"TENER: Adapting Transformer Encoder for Named Entity Recognition","code":"https://github.com/HIT-SCIR/ltp","n_code_links":6,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}}],"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":11,"rows_with_any_sample_ran":5,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":14,"n_unverified":52,"n_samples":66,"n_pointer_only_licence":5,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":21,"n_unverified":81,"n_samples":102,"n_pointer_only_licence":6,"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"}}}