Massively Multilingual Transfer for NER (2019-02-01T00:00:00.000000Z)

TL;DR

Evaluating on named entity recognition, it is shown that the proposed techniques for modulating the transfer are much more effective than strong baselines, including standard ensembling, and the unsupervised method rivals oracle selection of the single best individual model.

Abstract

In cross-lingual transfer, NLP models over one or more source languages are applied to a low-resource target language. While most prior work has used a single source model or a few carefully selected models, here we consider a “massive” setting with many such models. This setting raises the problem of poor transfer, particularly from distant languages. We propose two techniques for modulating the transfer, suitable for zero-shot or few-shot learning, respectively. Evaluating on named entity recognition, we show that our techniques are much more effective than strong baselines, including standard ensembling, and our unsupervised method rivals oracle selection of the single best individual model.

Authors

Trevor Cohn

7 papers

Afshin Rahimi

1 papers

Yuan Li

1 papers

TL;DR

Abstract

Authors

References50 items

Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsing

Low-resource named entity recognition via multi-source projection: Not quite there yet?

Distant Supervision from Disparate Sources for Low-Resource Part-of-Speech Tagging

Neural Cross-Lingual Named Entity Recognition with Minimal Resources

A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings

Building Named Entity Recognition Taggers via Parallel Corpora

Low-Resource Named Entity Recognition with Cross-lingual, Character-Level Neural Conditional Random Fields

Snorkel: Rapid Training Data Creation with Weak Supervision

Word Translation Without Parallel Data

Cheap Translation for Cross-Lingual Named Entity Recognition

Learning how to Active Learn: A Deep Reinforcement Learning Approach

Weakly Supervised Cross-Lingual Named Entity Recognition via Effective Annotation and Representation Projection

Model Transfer for Tagging Low-resource Languages using a Bilingual Dictionary

Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation

Phonologically Aware Neural Model for Named Entity Recognition in Low Resource Transfer Settings

Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser

Cross-Lingual Named Entity Recognition via Wikification

Enriching Word Vectors with Subword Information

Learning when to trust distant supervision: An application to low-resource POS tagging using cross-lingual projection

Learning Crosslingual Word Embeddings without Bilingual Corpora

Sequence-Level Knowledge Distillation

Neural Architectures for Named Entity Recognition

End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF

Many Languages, One Parser

Density-Driven Cross-Lingual Transfer of Dependency Parsers

Low Resource Dependency Parsing: Cross-lingual Parameter Sharing in a Neural Network Parser

Unsupervised Dependency Parsing with Transferring Distribution via Parallel Guidance and Entropy Regularization

Learning a Part-of-Speech Tagger from Two Hours of Annotation

Nudging the Envelope of Direct Transfer Methods for Multilingual Named Entity Recognition

Cross-lingual Word Clusters for Direct Transfer of Linguistic Structure

Bayesian Classifier Combination

Building a Multilingual Named Entity-Annotated Corpus Using Annotation Projection

Unsupervised Part-of-Speech Tagging with Bilingual Graph-Based Projections

The Multidimensional Wisdom of Crowds

Whose Vote Should Count More: Optimal Integration of Labels from Labelers of Unknown Expertise

Adding More Languages Improves Unsupervised Multilingual Part-of-Speech Tagging: a Bayesian Non-Parametric Approach

Mention Detection Crossing the Language Barrier

Bootstrapping parsers via syntactic projection across parallel texts

Introduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition

Introduction to the CoNLL-2002 Shared Task: Language-Independent Named Entity Recognition

Inducing Multilingual Text Analysis Tools via Robust Projection across Aligned Corpora

Maximum Likelihood Estimation of Observer Error‐Rates Using the EM Algorithm

Aggregating and Predicting Sequence Labels from Crowd Annotations

Cross-lingual Name Tagging and Linking for 282 Languages

Simple task-specific bilingual word embeddings

What Can We Get From 1000 Tokens? A Case Study of Multilingual POS Tagging For Resource-Poor Languages

Learning multilingual named entity recognition from Wikipedia

Eliminating Spammers and Ranking Annotators for Crowdsourced Labeling Tasks

Learning in graphical models , volume 89

supervised estimation of BEA prior ( § 2.2)

Field of Study

Venue Information

Name

Type

URL

Alternate Names