{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T04:31:42Z","timestamp":1780374702147,"version":"3.54.1"},"reference-count":33,"publisher":"MIT Press","license":[{"start":{"date-parts":[[2023,3,16]],"date-time":"2023-03-16T00:00:00Z","timestamp":1678924800000},"content-version":"vor","delay-in-days":74,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2023,3,14]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Most recent coreference resolution systems use search algorithms over possible spans to identify mentions and resolve coreference. We instead present a coreference resolution system that uses a text-to-text (seq2seq) paradigm to predict mentions and links jointly. We implement the coreference system as a transition system and use multilingual T5 as an underlying language model. We obtain state-of-the-art accuracy on the CoNLL-2012 datasets with 83.3 F1-score for English (a 2.3 higher F1-score than previous work [Dobrovolskii, 2021]) using only CoNLL data for training, 68.5 F1-score for Arabic (+4.1 higher than previous work), and 74.3 F1-score for Chinese (+5.3). In addition we use the SemEval-2010 data sets for experiments in the zero-shot setting, a few-shot setting, and supervised setting using all available training data. We obtain substantially higher zero-shot F1-scores for 3 out of 4 languages than previous approaches and significantly exceed previous supervised state-of-the-art results for all five tested languages. We provide the code and models as open source.1<\/jats:p>","DOI":"10.1162\/tacl_a_00543","type":"journal-article","created":{"date-parts":[[2023,3,16]],"date-time":"2023-03-16T14:36:49Z","timestamp":1678977409000},"page":"212-226","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":19,"title":["Coreference Resolution through a seq2seq Transition-Based System"],"prefix":"10.1162","volume":"11","author":[{"given":"Bernd","family":"Bohnet","sequence":"first","affiliation":[{"name":"Google Research, The Netherlands. bohnetbd@google.com"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chris","family":"Alberti","sequence":"additional","affiliation":[{"name":"Google Research, USA. chrisalberti@google.com"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Collins","sequence":"additional","affiliation":[{"name":"Google Research, USA. mjcollins@google.com"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2023,3,14]]},"reference":[{"key":"2023031614352446100_","first-page":"99","article-title":"Neural coreference resolution for Arabic","volume-title":"Proceedings of the Third Workshop on Computational Models of Reference, Anaphora and Coreference","author":"Aloraini","year":"2020"},{"key":"2023031614352446100_","first-page":"108","article-title":"TANL-1: Coreference resolution by parse analysis and similarity clustering","volume-title":"Proceedings of the 5th International Workshop on Semantic Evaluation","author":"Attardi","year":"2010"},{"key":"2023031614352446100_","doi-asserted-by":"crossref","first-page":"57","DOI":"10.18653\/v1\/2021.crac-1.6","article-title":"Lazy low-resource coreference resolution: A study on leveraging black-box translation tools","volume-title":"Proceedings of the Fourth Workshop on Computational Models of Reference, Anaphora and Coreference","author":"Bitew","year":"2021"},{"key":"2023031614352446100_","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"CoRR"},{"key":"2023031614352446100_","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)","author":"Devlin","year":"2019"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"7670","DOI":"10.18653\/v1\/2021.emnlp-main.605","article-title":"Word-level coreference resolution","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","author":"Dobrovolskii","year":"2021"},{"key":"2023031614352446100_","article-title":"Glam: Efficient scaling of language models with mixture-of-experts","author":"Nan","year":"2021","journal-title":"CoRR"},{"issue":"8","key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Computation"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1162\/tacl_a_00300","article-title":"SpanBERT: Improving pre-training by representing and predicting spans","volume":"8","author":"Joshi","year":"2020","journal-title":"Transactions of the Association for Computational Linguistics"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"5803","DOI":"10.18653\/v1\/D19-1588","article-title":"BERT for coreference resolution: Baselines and analysis","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Joshi","year":"2019"},{"key":"2023031614352446100_","volume-title":"Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition","author":"Jurafsky","year":"2021","edition":"third edition"},{"key":"2023031614352446100_","article-title":"Scaling laws for neural language models","author":"Kaplan","year":"2020"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"14","DOI":"10.18653\/v1\/2021.acl-short.3","article-title":"Coreference resolution without span representations","volume-title":"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)","author":"Kirstain","year":"2021"},{"key":"2023031614352446100_","first-page":"92","article-title":"SUCRE: A modular system for coreference resolution","volume-title":"Proceedings of the 5th International Workshop on Semantic Evaluation","author":"Kobdani","year":"2010"},{"key":"2023031614352446100_","first-page":"188","article-title":"End-to-end neural coreference resolution","volume-title":"Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing","author":"Lee","year":"2017"},{"key":"2023031614352446100_","first-page":"687","article-title":"Higher-order coreference resolution with coarse-to-fine inference","volume-title":"Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)","author":"Lee","year":"2018"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"94","DOI":"10.18653\/v1\/2021.crac-1.10","article-title":"Exploring pre-trained transformers and bilingual transfer learning for Arabic coreference resolution","volume-title":"Proceedings of the Fourth Workshop on Computational Models of Reference, Anaphora and Coreference","author":"Min","year":"2021"},{"key":"2023031614352446100_","first-page":"149","article-title":"An efficient algorithm for projective dependency parsing","volume-title":"Proceedings of the Eighth International Conference on Parsing Technologies","author":"Nivre","year":"2003"},{"issue":"4","key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1162\/coli.07-056-R1-07-027","article-title":"Algorithms for deterministic incremental dependency parsing","volume":"34","author":"Nivre","year":"2008","journal-title":"Computational Linguistics"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"2227","DOI":"10.18653\/v1\/N18-1202","article-title":"Deep contextualized word representations","volume-title":"Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)","author":"Peters","year":"2018"},{"key":"2023031614352446100_","first-page":"1","article-title":"CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes","volume-title":"Joint Conference on EMNLP and CoNLL - Shared Task","author":"Pradhan","year":"2012"},{"key":"2023031614352446100_","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","author":"Raffel","year":"2019","journal-title":"CoRR"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3115\/1621969.1621982","article-title":"SemEval-2010 task 1: Coreference resolution in multiple languages","volume-title":"Proceedings of the 5th International Workshop on Semantic Evaluation","author":"Recasens","year":"2010"},{"key":"2023031614352446100_","first-page":"155","article-title":"IMS HotCoref DE: A data-driven co-reference resolver for German","volume-title":"Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC\u201916)","author":"Roesiger","year":"2016"},{"key":"2023031614352446100_","first-page":"170","article-title":"Neural end-to-end coreference resolution for German in different domains","volume-title":"Proceedings of the 17th Conference on Natural Language Processing (KONVENS 2021)","author":"Schr\u00f6der","year":"2021"},{"key":"2023031614352446100_","article-title":"Lamda: Language models for dialog applications","author":"Thoppilan","year":"2022","journal-title":"CoRR"},{"key":"2023031614352446100_","first-page":"2129","article-title":"Limited memory incremental coreference resolution","volume-title":"Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers","author":"Webster","year":"2014"},{"key":"2023031614352446100_","doi-asserted-by":"publisher","first-page":"6953","DOI":"10.18653\/v1\/2020.acl-main.622","article-title":"CorefQA: Coreference resolution as query-based span prediction","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","author":"Wei","year":"2020"},{"key":"2023031614352446100_","doi-asserted-by":"crossref","first-page":"5241","DOI":"10.18653\/v1\/2021.emnlp-main.425","article-title":"Moving on from OntoNotes: Coreference resolution model transfer","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","author":"Xia","year":"2021"},{"key":"2023031614352446100_","doi-asserted-by":"crossref","first-page":"8617","DOI":"10.18653\/v1\/2020.emnlp-main.695","article-title":"Incremental neural coreference resolution in constant memory","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Xia","year":"2020"},{"key":"2023031614352446100_","first-page":"8527","article-title":"Revealing the myth of higher-order inference in coreference resolution","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Liyan","year":"2020"},{"key":"2023031614352446100_","first-page":"483","article-title":"mT5: A massively multilingual pre-trained text-to-text transformer","volume-title":"Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","author":"Xue","year":"2021"},{"key":"2023031614352446100_","first-page":"11","article-title":"A cluster ranking model for full anaphora resolution","volume-title":"Proceedings of the 12th Language Resources and Evaluation Conference","author":"Juntao","year":"2020"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/tacl_a_00543\/2074888\/tacl_a_00543.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/tacl_a_00543\/2074888\/tacl_a_00543.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,12,9]],"date-time":"2023-12-09T02:16:05Z","timestamp":1702088165000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/doi\/10.1162\/tacl_a_00543\/115243\/Coreference-Resolution-through-a-seq2seq"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023]]},"references-count":33,"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00543","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023]]},"published":{"date-parts":[[2023]]}}}