Transferring Coreference Resolvers with Posterior Regularization

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, pages 1427–1437, Beijing, China, July 26-31, 2015. c⃝2015 Association for Computational Linguistics Transferring Coreference Resolvers with Posterior Regularization Andr´e F. T. Martins∗† ∗Priberam Labs, Alameda D. Afonso Henriques, 41, 2o, 1000-123 Lisboa, Portugal †Instituto de Telecomunicac¸˜oes, Instituto Superior T´ecnico, 1049-001 Lisboa, Portugal atm@priberam.pt Abstract We propose a cross-lingual framework for learning coreference resolvers for resource-poor target languages, given a re- solver in a source language. Our method uses word-aligned bitext to project infor- mation from the source to the target. To handle task-specific costs, we propose a softmax-margin variant of posterior regu- larization, and we use it to achieve robust- ness to projection errors. We show empir- ically that this strategy outperforms com- petitive cross-lingual methods, such as delexicalized transfer with bilingual word embeddings, bitext direct projection, and vanilla posterior regularization. 1 Introduction The goal of coreference resolution is to find the mentions in text that refer to the same discourse entity. While early work focused primarily on En- glish (Soon et al., 2001; Ng and Cardie, 2002), efforts have been made toward multilingual sys- tems, this being addressed in recent shared tasks (Recasens et al., 2010; Pradhan et al., 2012). How- ever, the lack of annotated data hinders rapid sys- tem deployment for new languages. Unsupervised methods (Haghighi and Klein, 2007; Ng, 2008) and rule-based approaches (Raghunathan et al., 2010) avoid this data annotation bottleneck, but they often require complex generative models or expert linguistic knowledge. We propose cross-lingual coreference resolu- tion as a way of transferring information from a rich-resource language to build coreference re- solvers for languages with scarcer resources; as a testbed, we transfer from English to Spanish and to Brazilian Portuguese. We build upon the recent successes of cross-lingual learning in NLP, which proved quite effective in several structured predic- tion tasks, such as POS tagging (T¨ackstr¨om et al., 2013), named entity recognition (Wang and Man- ning, 2014), dependency parsing (McDonald et al., 2011), semantic role labeling (Titov and Kle- mentiev, 2012), and fine-grained opinion mining (Almeida et al., 2015). The potential of these tech- niques, however, has never been fully exploited in coreference resolution (despite some existing work, reviewed in §6, but none resulting in an end- to-end coreference resolver). We bridge this gap by proposing a simple learning-based method with weak supervision, based on posterior regularization (Ganchev et al., 2010). We adapt this framework to handle softmax-margin objective functions (Gimpel and Smith, 2010), leading to softmax-margin poste- rior regularization (§4). This step, while fairly simple, opens the door for incorporating task- specific cost functions, which are important to manage the precision/recall trade-offs in corefer- ence resolution systems. We show that the result- ing problem involves optimizing the difference of two cost-augmented log-partition functions, mak- ing a bridge with supervised systems based on la- tent coreference trees (Fernandes et al., 2012; Durrett and Klein, 2013), reviewed in §3. In- spired by this …

Content truncated to 500 words