Show simple item record

dc.contributor.authorThorne, Jen
dc.contributor.authorVlachos, Andreasen
dc.contributor.authorChristodoulopoulos, Cen
dc.contributor.authorMittal, Aen
dc.date.accessioned2020-06-11T23:30:22Z
dc.date.available2020-06-11T23:30:22Z
dc.date.issued2019-06en
dc.identifier.isbn9781950737130en
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/306712
dc.description.abstract© 2019 Association for Computational Linguistics The task of Natural Language Inference (NLI) is widely modeled as supervised sentence pair classification. While there has been a lot of work recently on generating explanations of the predictions of classifiers on a single piece of text, there have been no attempts to generate explanations of classifiers operating on pairs of sentences. In this paper, we show that it is possible to generate token-level explanations for NLI without the need for training data explicitly annotated for this purpose. We use a simple LSTM architecture and evaluate both LIME and Anchor explanations for this task. We compare these to a Multiple Instance Learning (MIL) method that uses thresholded attention make token-level predictions. The approach we present in this paper is a novel extension of zero-shot single-sentence tagging to sentence pairs for NLI. We conduct our experiments on the well-studied SNLI dataset that was recently augmented with manually annotation of the tokens that explain the entailment relation. We find that our white-box MIL-based method, while orders of magnitude faster, does not reach the same accuracy as the black-box methods.
dc.publisherAssociation for Computational Linguistics
dc.rightsAttribution 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.titleGenerating token-level explanations for natural language inferenceen
dc.typeConference Object
prism.endingPage969
prism.publicationDate2019en
prism.publicationNameProceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)en
prism.startingPage963
prism.volume1en
dc.identifier.doi10.17863/CAM.53800
dcterms.dateAccepted2019-02-22en
rioxxterms.versionofrecord10.18653/v1/N19-1101en
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2019-06en
dc.contributor.orcidVlachos, Andreas [0000-0003-2123-5071]
rioxxterms.typeConference Paper/Proceeding/Abstracten
pubs.conference-nameNAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologiesen
pubs.conference-start-date2019-06-02en


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Attribution 4.0 International
Except where otherwise noted, this item's licence is described as Attribution 4.0 International