Entity-Centric Coreference Resolution with Model Stacking

By Kevin Clark, Christopher D. Manning. 2015

209 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Management Science and Operations Research, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

In Natural Language Processing

Citations in the corpus

Toward Gender-Inclusive Coreference Resolution (2019, background): “2010], statistical [SfdS; Clark and Manning 2015] and neural [SfdN; Clark and Manning 2016] systems. Figure 4 shows the results. We can see that the average system accuracies mostly follow the same pattern as human agreement levels, though all are significantly lower than human results. In particula”
End-to-end Neural Coreference Resolution (2017, neutral): “…be categorized as (1) mention-pair classiﬁers (Ng and Cardie, 2002; Bengtson and Roth, 2008), (2) entity-level models (Haghighi and Klein, 2010; Clark and Manning, 2015, 2016b; Wiseman et al., 2016), (3) latent-tree models (Fernandes et al., 2012; Bj¨orkelund and Kuhn, 2014; Martschat and…”
Position-aware Attention and Supervised Data Improve Slot Filling (2017, methodology): “To further collect examples that are not annotated by LDC, we ﬁrst run annotations on the corpus using a combination of Stanford’s statistical coreference system (Clark and Manning, 2015) and the Illinois Wik-iﬁer (Ratinov et al., 2011).”
Higher-Order Coreference Resolution with Coarse-to-Fine Inference (2018, background): “Since global features are notoriously difﬁcult to deﬁne (Wiseman et al., 2016), they often depend heavily on existing pairwise features or architectures (Bj¨orkelund and Kuhn, 2014; Clark and Manning, 2015, 2016b).”
A Hierarchical Multi-task Approach for Learning Embeddings from Semantic Tasks (2018, background): “Using gold mentions is impossible in real settings so we also relax this condition leading to a more challenging task in which we make no use of external tools or metadata (such as speaker ID used by some systems (Clark and Manning 2015)).”
Improving Coreference Resolution by Learning Entity-Level Distributed Representations (2016, methodology): “We train our system using a learning-to-search algorithm similar to SEARN (Daumé III et al., 2009). Learning-to-search style algorithms have been employed to train coreference resolvers on trajectories of decisions similar to those that would be seen at test-time by Daumé et al. (2005), Ma et al.”
Visual Reference Resolution using Attention Memory for Visual Dialog (2017, methodology): “…attention α mem t and key k mem t are then retrieved from the attention memory using the computed addressing vector β t by This relevant attention retrieval allows the proposed model to resolve the visual reference by indirectly resolving coreferences [34–36] through the memory addressing process.”
Which Coreference Evaluation Metric Do You Trust? A Proposal for a Link-based Entity Aware Metric (2016, background): “Clark and Manning (2015)’s resolver is also among the state-of-the-art systems but we did not have access to their output.”
Neural Models for Reasoning over Multiple Mentions Using Coreference (2018, methodology): “For this dataset we used Stanford CoreNLP to extract coreferences (Clark and Manning, 2015), which achieved 0 .”
The KnowRef Coreference Corpus: Removing Gender and Number Cues for Difficult Pronominal Anaphora Resolution (2018, background): “, 2010) (Rule), Stanford’s statistical system (Clark and Manning, 2015) (Stat), Clark and Manning (2016)’s deep reinforcement learning system (Deep-RL), Martschat and Strube (2015)’s latent tree model (Latent), and Lee et al.”
MSnet: A BERT-based Network for Gendered Pronoun Resolution (2019, background): “Up to now, various models for coreference resolution have been proposed, and they can be generally categorized as (1) mention-pair classifier model (Webster and Nothman, 2016), (2) entity-centric model (Clark and Manning, 2015), (3) ranking model (Lee et al.”
Resolving Gendered Ambiguous Pronouns with BERT (2019, unknown): “They run 4 state-of-the-art coreference resolution models (Lee et al., 2013; Clark and Manning, 2015; Wiseman et al., 2016; Lee et al., 2017) on the OntoNotes and GAP datasets reporting F1 scores separately for masculine and feminine pronoun-named entity pairs (metrics M and F in the paper).”
Evaluation of named entity coreference (2019, methodology): “…evaluate the Stanford coreference system, with its deterministic (Raghunathan et al., 2010; Lee et al., 2011; Recasens et al., 2013), statistical (Clark and Manning, 2015) and neural (Clark and Manning, 2016a,b) versions, and the neural end-to-end systems of (Lee et al., 2017) and (Lee et al.,…”
BERT for Coreference Resolution: Baselines and Analysis (2019, unknown): “ttp://github.com/kentonl/e2e-coref/ 4https://github.com/google-research/bert MUC B3 CEAF φ4 P R F1 P R F1 P R F1 Avg. F1 Martschat and Strube (2015) 76.7 68.1 72.2 66.1 54.2 59.6 59.5 52.3 55.7 62.5 (Clark and Manning, 2015) 76.1 69.4 72.6 65.6 56.0 60.4 59.4 53.0 56.0 63.0 (Wiseman et al., 2015) 76”
CorefQA: Coreference Resolution as Query-based Span Prediction (2019, methodology): “Earlier neural-based models (Wiseman et al., 2016; Clark and Manning, 2015, 2016) rely on parsers and hand-engineered mention proposal algorithms.”
Gendered Pronoun Resolution using BERT and an Extractive Question Answering Formulation (2019, background): “The authors show that even the best of the baselines such as (Clark and Manning, 2015), (Wiseman et al., 2016), (Lee et al., 2017) achieve an F1 score of just 66.9% on this dataset.”
Learning Global Features for Coreference Resolution (2016, background): “Given a sequence (xn)Nn=1 of (intra-document) mentions – that is, syntactic units that can refer or be referred to – coreference resolution involves partitioning (xn) into a sequence of clusters (X(m))Mm=1 such that all the mentions in any particular cluster 994 X(m) refer to the same underlying…”