Repository logo
 

Zero-shot Sequence Labeling: Transferring Knowledge from Sentences to Tokens

Accepted version
Peer-reviewed

Loading...
Thumbnail Image

Type

Conference Object

Change log

Authors

Rei, M 
Søgaard, Anders 

Abstract

Can attention- or gradient-based visualization techniques be used to infer token-level labels for binary sequence tagging problems, using networks trained only on sentence-level labels? We construct a neural network architecture based on soft attention, train it as a binary sentence classifier and evaluate against token-level annotation on four different datasets. Inferring token labels from a network provides a method for quantitatively evaluating what the model is learning, along with generating useful feedback in assistance systems. Our results indicate that attention-based methods are able to predict token-level labels more accurately, compared to gradient-based methods, sometimes even rivaling the supervised oracle network.

Description

Keywords

Journal Title

Conference Name

The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Journal ISSN

Volume Title

Publisher

Sponsorship
Cambridge Assessment (unknown)
ERC Nvidia