Narrowing the Modeling Gap: A Cluster-Ranking Approach to Coreference Resolution

Rahman, Altaf; Ng, Vincent

doi:10.1613/jair.3120

Cited by 37 publications

(49 citation statements)

References 58 publications

(75 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Only Scoring is the special case of our Prune-andScore approach where we employ only the scoring function. This corresponds to existing incremental approaches (Daumé III, 2006;Rahman and Ng, 2011b). We report the best published results for CPL 3 M system, Easy-first, and Fernandes et al, 2012.…”

Section: Resultsmentioning

confidence: 55%

“…The indicator feature will be 1 for the NEW action and 0 for all other actions.We have a total of 140 features: 90 mention pair features; 49 entity pair features; and one NEW indicator feature. We believe that our approach can benefit from employing features of the mention for the NEW action (Rahman and Ng, 2011b;. However, we were constrained by the Reconcile system and could not leverage these features for the NEW action.…”

Section: Experiments and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Prune-and-Score: Learning for Greedy Coreference Resolution

Doppa

Orr³

et al. 2014

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

We propose a novel search-based approach for greedy coreference resolution, where the mentions are processed in order and added to previous coreference clusters. Our method is distinguished by the use of two functions to make each coreference decision: a pruning function that prunes bad coreference decisions from further consideration, and a scoring function that then selects the best among the remaining decisions. Our framework reduces learning of these functions to rank learning, which helps leverage powerful off-the-shelf rank-learners. We show that our Prune-and-Score approach is superior to using a single scoring function to make both decisions and outperforms several state-of-the-art approaches on multiple benchmark corpora including OntoNotes.

show abstract

Section: Resultsmentioning

confidence: 55%

Section: Experiments and Resultsmentioning

confidence: 99%

Prune-and-Score: Learning for Greedy Coreference Resolution

Doppa

Orr³

et al. 2014

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

show abstract

“…Several works have explored using non-local entity-level features in mention-entity models that assign a single mention to a (partially completed) cluster (Luo et al, 2004;Yang et al, 2008;Rahman and Ng, 2011). Our system, however, builds clusters incrementally through merge operations, and so can operate in an easy-first fashion.…”

Section: Related Workmentioning

confidence: 99%

Entity-Centric Coreference Resolution with Model Stacking

Clark

Manning

2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

140

155

View full text Add to dashboard Cite

Mention pair models that predict whether or not two mentions are coreferent have historically been very effective for coreference resolution, but do not make use of entity-level information. However, we show that the scores produced by such models can be aggregated to define powerful entity-level features between clusters of mentions. Using these features, we train an entity-centric coreference system that learns an effective policy for building up coreference chains incrementally. The mention pair scores are also used to prune the search space the system works in, allowing for efficient training with an exact loss function. We evaluate our system on the English portion of the 2012 CoNLL Shared Task dataset and show that it improves over the current state of the art.

show abstract

“…Lee et al (2011) proposed a competitive non-learned sieve-based method, which constructs clusters by aglomerating mentions in a greedy manner. Entity-centric models define scores for the entire entity clusters (Culotta et al, 2007;Haghighi and Klein, 2010;Rahman and Ng, 2011) and seek the set of entities that optimize the sum of scores; this can also be promoted in a decentralized manner . Pairwise models (Bengtson and Roth, 2008;Finkel et al, 2008;Versley et al, 2008), on the other hand, define scores for each pair of mentions to be coreferent, and define the clusters as the transitive closure of these pairwise relations.…”

Section: Coreference Resolutionmentioning

confidence: 99%

A Joint Model for Quotation Attribution and Coreference Resolution

Almeida

Martins

2014

Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics

View full text Add to dashboard Cite

We address the problem of automatically attributing quotations to speakers, which has great relevance in text mining and media monitoring applications. While current systems report high accuracies for this task, they either work at mentionlevel (getting credit for detecting uninformative mentions such as pronouns), or assume the coreferent mentions have been detected beforehand; the inaccuracies in this preprocessing step may lead to error propagation. In this paper, we introduce a joint model for entity-level quotation attribution and coreference resolution, exploiting correlations between the two tasks. We design an evaluation metric for attribution that captures all speakers' mentions. We present results showing that both tasks benefit from being treated jointly.

show abstract

Narrowing the Modeling Gap: A Cluster-Ranking Approach to Coreference Resolution

Cited by 37 publications

References 58 publications

Prune-and-Score: Learning for Greedy Coreference Resolution

Prune-and-Score: Learning for Greedy Coreference Resolution

Entity-Centric Coreference Resolution with Model Stacking

A Joint Model for Quotation Attribution and Coreference Resolution

Contact Info

Product

Resources

About