Back to Nlp Progress

Coreference resolution

english/coreference_resolution.md

0.34.0 KB
Original Source

Coreference resolution

Coreference resolution is the task of clustering mentions in text that refer to the same underlying real world entities.

Example:

               +-----------+
               |           |
I voted for Obama because he was most aligned with my values", she said.
 |                                                 |            |
 +-------------------------------------------------+------------+

"I", "my", and "she" belong to the same cluster and "Obama" and "he" belong to the same cluster.

CoNLL 2012

Experiments are conducted on the data of the CoNLL-2012 shared task, which uses OntoNotes coreference annotations. Papers report the precision, recall, and F1 of the MUC, B3, and CEAFφ4 metrics using the official CoNLL-2012 evaluation scripts. The main evaluation metric is the average F1 of the three metrics.

ModelAvg F1Paper / SourceCode
wl-coref + RoBERTa81.0Word-Level Coreference ResolutionOfficial
s2e+Longformer-Large80.3Coreference Resolution without Span RepresentationsOfficial
Xu et al. (2020)80.2Revealing the Myth of Higher-Order Inference in Coreference ResolutionOfficial
Joshi et al. (2019)<sup>1</sup>79.6SpanBERT: Improving Pre-training by Representing and Predicting SpansOfficial
Joshi et al. (2019)<sup>2</sup>76.9BERT for Coreference Resolution: Baselines and AnalysisOfficial
Kantor and Globerson (2019)76.6Coreference Resolution with Entity EqualizationOfficial
Fei et al. (2019)73.8End-to-end Deep Reinforcement Learning Based Coreference Resolution
(Lee et al., 2017)+ELMo (Peters et al., 2018)+coarse-to-fine & second-order inference (Lee et al., 2018)73.0Higher-order Coreference Resolution with Coarse-to-fine InferenceOfficial
(Lee et al., 2017)+ELMo (Peters et al., 2018)70.4Deep contextualized word representations
Lee et al. (2017)67.2End-to-end Neural Coreference Resolution

<a name="myfootnote1">[1]</a> Joshi et al. (2019): (Lee et al., 2017)+coarse-to-fine & second-order inference (Lee et al., 2018)+SpanBERT (Joshi et al., 2019)

<a name="myfootnote2">[2]</a> Joshi et al. (2019): (Lee et al., 2017)+coarse-to-fine & second-order inference (Lee et al., 2018)+BERT (Devlin et al., 2019)

Gendered Ambiguous Pronoun Resolution

Experiments are conducted on GAP dataset. Metrics used are F1 score on Masculine (M) and Feminine (F) examples, Overall, and a Bias factor calculated as F / M.

ModelOverall F1Masculine F1 (M)Feminine F1 (F)Bias (F/M)Paper / SourceCode
Attree et al. (2019)92.594.091.10.97Gendered Ambiguous Pronouns Shared Task: Boosting Model Confidence by Evidence PoolingGREP
Chada et al. (2019)90.290.989.50.98Gendered Pronoun Resolution using BERT and an extractive question answering formulationCorefQA

Go back to the README