--- title: "TANL-1: Coreference Resolution by Parse — Attardi 2010" description: "This paper presents TANL-1, the University of Pisa system for the SemEval-2010 shared task on coreference resolution across several languages." authors: ["Giuseppe Attardi", "Maria Simi", "Stefano Dei Rossi"] year: 2010 venue: "Proceedings of the 5th International Workshop on Semantic Evaluation" source: "https://aclanthology.org/S10-1022/" licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/papers/attardi-2010-tanl-1-coreference-resolution-parse/" --- [Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/attardi-2010-tanl-1-coreference-resolution-parse/) [home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2010](https://md.research.elamur.ai/free/papers/2010/) / TANL-1: Coreference Resolution by Parse Analysis and Similarity Clustering # TANL-1: Coreference Resolution by Parse Analysis and Similarity Clustering **Authors:** [Giuseppe Attardi](https://md.research.elamur.ai/free/authors/giuseppe-attardi/), [Maria Simi](https://md.research.elamur.ai/free/authors/maria-simi/), Stefano Dei Rossi · **Year:** 2010 · **Venue:** Proceedings of the 5th International Workshop on Semantic Evaluation · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/) **Links:** [Publisher's page](https://aclanthology.org/S10-1022/) · [PDF](https://aclanthology.org/S10-1022.pdf) · [OpenAlex](https://openalex.org/W2251809029) **Citations:** 17 citations (openalex-snapshot · 2026-10-14) · [9](#cited-by) in our library, of the 140673 papers whose bibliographies we read ## Summary This paper presents TANL-1, the University of Pisa system for the SemEval-2010 shared task on coreference resolution across several languages. The authors split the task into two stages. - **Mention detection.** The system inspects dependency trees, using gold HEAD/DEPREL annotations when present and predicted parser output otherwise. Likely nominal heads are found with minimal language-specific knowledge, the leaves of their subtrees become mention candidates, and named entities from the NE column are merged in. - **Entity creation.** A Maximum Entropy pairwise classifier estimates whether two mentions corefer, using lexical, distance, syntactic, frequency, NE-type and pronoun features. A greedy best-first procedure then attaches each mention to the most probable earlier entity, provided the probability exceeds a threshold. The system was evaluated in both gold and regular settings for Catalan, English, German and Spanish. It reached the top score among participants on several measures, including the best B3 for German in the gold setting. The work shows that a simple parse-driven pipeline can be competitive. It also shows that inconsistent annotation conventions across treebanks seriously degrade mention boundary detection. ## Key points - **Training and features.** The Maximum Entropy classifier is implemented in Python (Le, 2004) and trained with GIS for 600 iterations, which outperformed L-BFGS. Training took between 1 minute (German) and 8 minutes (Italian). Features include string match, prefix/suffix, acronym, edit distance, sentence/token/mention distance, head POS, mention counts, NE type, and pronoun gender, number and type. - **Mention detection F1 in the regular task.** Scores were high where dependency and NE annotations were consistent: Catalan 82.7 and Spanish 83.1. German dropped to 59.2 because punctuation was attached inside noun-phrase subtrees. - **English and Italian annotation problems.** In English, dependency and named-entity annotations followed different conventions, so boundaries disagreed. In Italian, named entities sometimes included prepositions, and some mentions crossed sentence boundaries. - **Italian submission.** POS and morphology were re-tagged with a modified Hunpos tagger, which made it an open-task submission. Errors in the test data delayed the run, so only unofficial results are reported. - **Gold setting.** The same parse-based detection was applied instead of using the provided mention boundaries. This explains the slight loss in mention accuracy, for example about 98.4 for Catalan and Spanish. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/) Topics: [Named entity recognition](https://md.research.elamur.ai/free/topics/named-entity-recognition/) · [Coreference resolution](https://md.research.elamur.ai/free/topics/coreference-resolution/) ## Cited by Cited by 9 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 5 not yet. - [Coreference Resolution through a seq2seq Transition-Based System](https://md.research.elamur.ai/free/papers/bohnet-2023-coreference-resolution-through-seq2seq-transition/) - [Transferring Coreference Resolvers with Posterior Regularization](https://md.research.elamur.ai/free/papers/martins-2015-transferring-coreference-resolvers-posterior-regularization/) - [Translation-Based Projection for Multilingual Coreference Resolution](https://md.research.elamur.ai/free/papers/rahman-2012-translation-based-projection-multilingual-coreference/) - [IMS HotCoref DE: A Data-driven Co-reference Resolver for German](https://md.research.elamur.ai/free/papers/roesiger-2016-ims-hotcoref-de-data-driven/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{attardi2010tanl, title = {{TANL-1: Coreference Resolution by Parse Analysis and Similarity Clustering}}, author = {Attardi, Giuseppe and Simi, Maria and Rossi, Stefano Dei}, booktitle = {Proceedings of the 5th International Workshop on Semantic Evaluation}, year = {2010}, url = {https://aclanthology.org/S10-1022/} } ``` - [← previous TAGME: On-the-fly Annotation of Short Text Fragments (by Wikipedia Entities)](https://md.research.elamur.ai/free/papers/ferragina-2010-tagme-fly-annotation-short/) - [next → The CoNLL-2010 Shared Task: Learning to Detect Hedges and their Scope in Natural Language Text](https://md.research.elamur.ai/free/papers/farkas-2010-conll-2010-shared-task-learning/) - [↑ back to papers of 2010](https://md.research.elamur.ai/free/papers/2010/) ## References 6 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. G. Attardi et al., 2009. Tanl (Text Analytics and Natural Language Processing). SemaWiki project: [http://medialab.di.unipi.it/wiki/SemaWiki](http://medialab.di.unipi.it/wiki/SemaWiki). [↗](http://medialab.di.unipi.it/wiki/SemaWiki) 2. P. Halácsy, A. Kornai, and C. Oravecz, 2007. Hun-Pos: an open source trigram tagger. Proceedings of the ACL 2007, Prague. [find](https://scholar.google.com/scholar?q=P%2E+Hal%C3%A1csy%2C+A%2E+Kornai%2C+and+C%2E+Oravecz%2C+2007%2E+Hun%2DPos%3A+an+open+source+trigram+tagger%2E+Proceedings+of+the+ACL+2007%2C+Prague%2E) 3. Z. Le, Maximum Entropy Modeling Toolkit for Pytho and C++, Reference Manual. [find](https://scholar.google.com/scholar?q=Z%2E+Le%2C+Maximum+Entropy+Modeling+Toolkit+for+Pytho+and+C%2B%2B%2C+Reference+Manual%2E) 4. X. Luo, A. Ittycheriah, H. Jing, N. Kambhatla & S. Roukos. 2004. A Mention-Synchronous Coreference Resolution Algorithm Based on the Bell Tree. Proceedings of the ACL 2004, Barcelona. [find](https://scholar.google.com/scholar?q=X%2E+Luo%2C+A%2E+Ittycheriah%2C+H%2E+Jing%2C+N%2E+Kambhatla+%26+S%2E+Roukos%2E+2004%2E+A+Mention%2DSynchronous+Coreference+Resolution+Algorithm+Based+on+the+Bell+Tree%2E+Proceedings+of+the+ACL+2004%2C+Barcelona%2E) [our card](https://md.research.elamur.ai/free/papers/luo-2004-mention-synchronous-coreference-resolution-algorithm/) 5. V. Ng, Machine Learning for Coreference Resolution: From Local Classification to Global Ranking, Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL), Ann Arbor, MI, June 2005, pp. 157-164. [find](https://scholar.google.com/scholar?q=V%2E+Ng%2C+Machine+Learning+for+Coreference+Resolution%3A+From+Local+Classification+to+Global+Ranking%2C+Proceedings+of+the+43rd+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28ACL%29%2C+Ann+Arbo) 6. M. Recasens, L. Màrquez, E. Sapena, M. A. Martí, M. Taulé, V. Hoste, M. Poesio and Y. Versley, SemEval-2010 Task 1: Coreference resolution in multiple languages, in Proceedings of the 5th International Workshop on Semantic Evaluations (SemEval-2010), Uppsala, Sweden, 2010. [find](https://scholar.google.com/scholar?q=M%2E+Recasens%2C+L%2E+M%C3%A0rquez%2C+E%2E+Sapena%2C+M%2E+A%2E+Mart%C3%AD%2C+M%2E+Taul%C3%A9%2C+V%2E+Hoste%2C+M%2E+Poesio+and+Y%2E+Versley%2C+SemEval%2D2010+Task+1%3A+Coreference+resolution+in+multiple+languages%2C+in+Proceedings+of+the+5th+Internationa) [our card](https://md.research.elamur.ai/free/papers/recasens-2009-semeval-2010-task-1-coreference/)