TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Coreference Resolution	CoNLL 2012	c2f-coref + BERT-large	Avg F1	76.9	# 10
Coreference Resolution	OntoNotes	BERT-large	F1	76.9	# 14
Coreference Resolution	OntoNotes	BERT-base	F1	73.9	# 16

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/bert-for-coreference-resolution-baselines-and/coreference-resolution-on-conll-2012)](https://paperswithcode.com/sota/coreference-resolution-on-conll-2012?p=bert-for-coreference-resolution-baselines-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/bert-for-coreference-resolution-baselines-and/coreference-resolution-on-ontonotes)](https://paperswithcode.com/sota/coreference-resolution-on-ontonotes?p=bert-for-coreference-resolution-baselines-and)`

BERT for Coreference Resolution: Baselines and Analysis

IJCNLP 2019 · Mandar Joshi, Omer Levy, Daniel S. Weld, Luke Zettlemoyer ·

We apply BERT to coreference resolution, achieving strong improvements on the OntoNotes (+3.9 F1) and GAP (+11.5 F1) benchmarks. A qualitative analysis of model predictions indicates that, compared to ELMo and BERT-base, BERT-large is particularly better at distinguishing between related but distinct entities (e.g., President and CEO). However, there is still room for improvement in modeling document-level context, conversations, and mention paraphrasing. Our code and models are publicly available.

PDF Abstract IJCNLP 2019 PDF IJCNLP 2019 Abstract

Code

Add Remove Mark official

mandarjoshi90/coref official

↳ Quickstart in

Colab

434

wooseok-AI/Korean_e2e_CR_BERT

↳ Quickstart in

Colab

Tasks

Add Remove

Coreference Resolution

Datasets

OntoNotes 5.0 CoNLL

GAP Coreference Dataset CoNLL-2012

Results from the Paper

Edit

Ranked #10 on Coreference Resolution on CoNLL 2012 (using extra training data)

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Coreference Resolution	CoNLL 2012	c2f-coref + BERT-large	Avg F1	76.9	# 10	Compare
Coreference Resolution	OntoNotes	BERT-large	F1	76.9	# 14	Compare
Coreference Resolution	OntoNotes	BERT-base	F1	73.9	# 16	Compare

Methods

Add Remove

Adam • Attention Dropout • BERT • BiLSTM • Dense Connections • Dropout • ELMo • GELU • Layer Normalization • Linear Layer • Linear Warmup With Linear Decay • LSTM • Multi-Head Attention • Residual Connection • Scaled Dot-Product Attention • Sigmoid Activation • Softmax • Tanh Activation • Weight Decay • WordPiece

Edit Social Preview

BERT for Coreference Resolution: Baselines and Analysis

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove