Domain Adaptation with Active Learning for Coreference Resolution

Zhao, Shanheng; Ng, Hwee Tou

doi:10.3115/v1/w14-1104

Cited by 10 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Domain adaptation in coreference resolution has been discussed often, both in the context of multiple text types within standard reference corpora (e.g. conversation, newswire and Web subcorpora in datasets such as the ACE corpus, see Yang et al 2012) or novel domains that are not included in most reference corpora, such as Biomedical NLP (Apostolova et al 2012, Zhao & Ng 2014. Such studies suggest a genre or text type effect for coreference; sentence type effects, by contrast, have not yet been studied.…”

Section: Methodsmentioning

confidence: 99%

Different Flavors of GUM: Evaluating Genre and Sentence Type Effects on Multilayer Corpus Annotation Quality

Zeldes

Simonson

2016

Proceedings of the 10th Linguistic Annotation Workshop Held In Conjunction With ACL 2016 (LAW-X 2016)

View full text Add to dashboard Cite

Genre and domain are well known covariates of both manual and automatic annotation quality. Comparatively less is known about the effect of sentence types, such as imperatives, questions or fragments, and how they interact with text type effects. Using mixed effects models, we evaluate the relative influence of genre and sentence types on automatic and manual annotation quality for three related tasks in English data: POS tagging, dependency parsing and coreference resolution. For the latter task, we also develop a new metric for the evaluation of individual regions of coreference annotation. Our results show that while there are substantial differences between manual and automatic annotation in each task, sentence type is generally more important than genre in predicting errors within our data.

show abstract

Section: Methodsmentioning

confidence: 99%

Different Flavors of GUM: Evaluating Genre and Sentence Type Effects on Multilayer Corpus Annotation Quality

Zeldes

Simonson

2016

Proceedings of the 10th Linguistic Annotation Workshop Held In Conjunction With ACL 2016 (LAW-X 2016)

View full text Add to dashboard Cite

show abstract

“…Yang et al (2012) adapts a model trained on the MUC-6 and ACE 2005 datasets to the biomedical domain using an active learning approach, applying data augmentation and pruning techniques. Zhao and Ng (2014) propose a feature-based active learning method to learn cross-domain knowledge. Unlike these works, we take advantage of the modern expressive power of the SpanBERT representation.…”

Section: Metricmentioning

confidence: 99%

Improving Span Representation for Domain-adapted Coreference Resolution

Gandhi¹,

Field²,

Tsvetkov³

2021

Preprint

View full text Add to dashboard Cite

Recent work has shown fine-tuning neural coreference models can produce strong performance when adapting to different domains. However, at the same time, this can require a large amount of annotated target examples. In this work, we focus on supervised domain adaptation for clinical notes, proposing the use of concept knowledge to more efficiently adapt coreference models to a new domain. We develop methods to improve the span representations via (1) a retrofitting loss to incentivize span representations to satisfy a knowledgebased distance function and (2) a scaffolding loss to guide the recovery of knowledge from the span representation. By integrating these losses, our model is able to improve our baseline precision and F-1 score. In particular, we show that incorporating knowledge with endto-end coreference models results in better performance on the most challenging, domainspecific spans 1 .

show abstract

“…Meanwhile, models for coreference resolution have improved due to neural architectures with millions of parameters and the emergence of pretrained encoders. However, model generalization across domains has always been a challenge (Yang et al, 2012;Zhao and Ng, 2014;Poot and van Cranenburgh, 2020;Aktaş et al, 2020). Since these models are usually engineered for a single dataset, they capture idiosyncrasies inherent in that dataset.…”

Section: Introductionmentioning

confidence: 99%

On Generalization in Coreference Resolution

Toshniwal¹,

Xia²,

Wiseman³

et al. 2021

Proceedings of the Fourth Workshop on Computational Models of Reference, Anaphora and Coreference

View full text Add to dashboard Cite

While coreference resolution is defined independently of dataset domain, most models for performing coreference resolution do not transfer well to unseen domains. We consolidate a set of 8 coreference resolution datasets targeting different domains to evaluate the offthe-shelf performance of models. We then mix three datasets for training; even though their domain, annotation guidelines, and metadata differ, we propose a method for jointly training a single model on this heterogeneous data mixture by using data augmentation to account for annotation differences and sampling to balance the data quantities. We find that in a zeroshot setting, models trained on a single dataset transfer poorly while joint training yields improved overall performance, leading to better generalization in coreference resolution models. This work contributes a new benchmark for robust coreference resolution and multiple new state-of-the-art results. 1

show abstract

Domain Adaptation with Active Learning for Coreference Resolution

Cited by 10 publications

References 15 publications

Different Flavors of GUM: Evaluating Genre and Sentence Type Effects on Multilayer Corpus Annotation Quality

Different Flavors of GUM: Evaluating Genre and Sentence Type Effects on Multilayer Corpus Annotation Quality

Improving Span Representation for Domain-adapted Coreference Resolution

On Generalization in Coreference Resolution

Contact Info

Product

Resources

About