{"url":"/sota/relation-extraction-on-ace-2004","task":{"name":"Relation Extraction","url":"/task/relation-extraction","note":null},"dataset":{"name":"ACE 2004","url":"/dataset/ace-2004"},"category":"Natural Language Processing","categories":["Natural Language Processing"],"category_note":null,"description":"**Relation Extraction** is the task of predicting attributes and relations for entities in a sentence. For example, given a sentence “Barack Obama was born in Honolulu, Hawaii.”, a relation classifier aims at predicting the relation of “bornInCity”. Relation Extraction is the key component for building relation knowledge graphs, and it is of crucial significance to natural language processing applications such as structured search, sentiment analysis, question answering, and summarization.\r\n\r\n\r\n<span class=\"description-source\">Source: [Deep Residual Learning for Weakly-Supervised Relation Extraction ](https://arxiv.org/abs/1707.08866)</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":["RE+ Micro F1","RE Micro F1","NER Micro F1","Cross Sentence"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"RE+ Micro F1":"higher","RE Micro F1":"higher","NER Micro F1":"higher","Cross Sentence":null}},"counts":{"rows":11,"rows_with_code":9,"rows_with_paper_page":11,"rows_dated":11,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"PL-Marker","metrics":{"Cross Sentence":"Yes","NER Micro F1":"90.4","RE Micro F1":"69.7","RE+ Micro F1":"66.5"},"uses_additional_data":false,"paper_date":"2021-09-13","paper":"/paper/pack-together-entity-and-relation-extraction","paper_url":"https://arxiv.org/abs/2109.06067v5","paper_title":"Packed Levitated Marker for Entity and Relation Extraction","code":"https://github.com/tomaarsen/spanmarkerner","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":3,"n_samples":9,"n_pointer_only_licence":1}},{"rank_in_archive_order":2,"model":"PFN","metrics":{"Cross Sentence":"No","NER Micro F1":"89.3","RE+ Micro F1":"62.5"},"uses_additional_data":false,"paper_date":"2021-08-27","paper":"/paper/a-partition-filter-network-for-joint-entity","paper_url":"https://arxiv.org/abs/2108.12202v8","paper_title":"A Partition Filter Network for Joint Entity and Relation Extraction","code":"https://github.com/Coopercoppers/PFN","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"Ours: cross-sentence ALB","metrics":{"Cross Sentence":"Yes","NER Micro F1":"90.3","RE Micro F1":"66.1","RE+ Micro F1":"62.2"},"uses_additional_data":false,"paper_date":"2020-10-24","paper":"/paper/a-frustratingly-easy-approach-for-joint","paper_url":"https://arxiv.org/abs/2010.12812v2","paper_title":"A Frustratingly Easy Approach for Entity and Relation Extraction","code":"https://github.com/princeton-nlp/PURE","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":1}},{"rank_in_archive_order":4,"model":"Table-Sequence","metrics":{"Cross Sentence":"No","NER Micro F1":"88.6","RE Micro F1":"63.3","RE+ Micro F1":"59.6"},"uses_additional_data":false,"paper_date":"2020-10-08","paper":"/paper/two-are-better-than-one-joint-entity-and","paper_url":"https://arxiv.org/abs/2010.03851v1","paper_title":"Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders","code":"https://github.com/LorrinWWW/two-are-better-than-one","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":5,"model":"Multi-turn QA","metrics":{"Cross Sentence":"No","NER Micro F1":"83.6","RE+ Micro F1":"49.4"},"uses_additional_data":false,"paper_date":"2019-05-14","paper":"/paper/entity-relation-extraction-as-multi-turn","paper_url":"https://arxiv.org/abs/1905.05529v4","paper_title":"Entity-Relation Extraction as Multi-Turn Question Answering","code":"https://github.com/ShannonAI/Entity-Relation-As-Multi-Turn-QA","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"SPTree","metrics":{"Cross Sentence":"No","NER Micro F1":"81.8","RE+ Micro F1":"48.4"},"uses_additional_data":false,"paper_date":"2016-01-05","paper":"/paper/end-to-end-relation-extraction-using-lstms-on","paper_url":"http://arxiv.org/abs/1601.00770v3","paper_title":"End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures","code":"https://github.com/tticoin/LSTM-ER","n_code_links":2,"syntology":null},{"rank_in_archive_order":7,"model":"multi-head + AT","metrics":{"Cross Sentence":"No","NER Micro F1":"81.64","RE+ Micro F1":"47.45"},"uses_additional_data":false,"paper_date":"2018-08-21","paper":"/paper/adversarial-training-for-multi-context-joint","paper_url":"http://arxiv.org/abs/1808.06876v3","paper_title":"Adversarial training for multi-context joint entity and relation extraction","code":"https://github.com/bekou/multihead_joint_entity_relation_extraction","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"multi-head","metrics":{"Cross Sentence":"No","NER Micro F1":"81.16","RE+ Micro F1":"47.14"},"uses_additional_data":false,"paper_date":"2018-04-20","paper":"/paper/joint-entity-recognition-and-relation","paper_url":"http://arxiv.org/abs/1804.07847v3","paper_title":"Joint entity recognition and relation extraction as a multi-head selection problem","code":"https://github.com/bekou/multihead_joint_entity_relation_extraction","n_code_links":6,"syntology":null},{"rank_in_archive_order":9,"model":"Attention","metrics":{"Cross Sentence":"No","NER Micro F1":"79.6","RE+ Micro F1":"45.7"},"uses_additional_data":false,"paper_date":"2017-07-01","paper":"/paper/going-out-on-a-limb-joint-extraction-of","paper_url":"https://aclanthology.org/P17-1085","paper_title":"Going out on a limb: Joint Extraction of Entity Mentions and Relations without Dependency Trees","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"Joint w/ Global","metrics":{"Cross Sentence":"No","NER Micro F1":"79.7","RE+ Micro F1":"45.3"},"uses_additional_data":false,"paper_date":"2014-06-01","paper":"/paper/incremental-joint-extraction-of-entity","paper_url":"https://aclanthology.org/P14-1038","paper_title":"Incremental Joint Extraction of Entity Mentions and Relations","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"DyGIE","metrics":{"Cross Sentence":"Yes","NER Micro F1":"87.4","RE Micro F1":"59.7"},"uses_additional_data":false,"paper_date":"2019-04-05","paper":"/paper/a-general-framework-for-information","paper_url":"http://arxiv.org/abs/1904.03296v1","paper_title":"A General Framework for Information Extraction using Dynamic Span Graphs","code":"https://github.com/luanyi/DyGIE","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 7,081 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":7081,"papers_extracted_not_yet_verified":217,"boards_without_verdict":27,"papers_not_yet_extracted":2325},"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":4,"rows_with_any_sample_ran":3,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":3,"samples_over_distinct_papers":{"n_ran":14,"n_unverified":9,"n_samples":23,"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":14,"n_unverified":9,"n_samples":23,"n_pointer_only_licence":5,"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"}}}