GrammarSHAP: An Efficient Model-Agnostic and Structure-Aware NLP Explainer

Edoardo Mosca, Defne Demitürk, Luca Mülln, Fabio Raffagnato, Georg Groh

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Interpreting NLP models is fundamental for their development as it can shed light on hidden properties and unexpected behaviors. However, while transformer architectures exploit contextual information to enhance their predictive capabilities, most of the available methods to explain such predictions only provide importance scores at the word level. This work addresses the lack of feature attribution approaches that also take into account the sentence structure. We extend the SHAP framework by proposing GrammarSHAP—a model-agnostic explainer leveraging the sentence’s constituency parsing to generate hierarchical importance scores.

Original languageEnglish
Title of host publicationLNLS 2022 - 1st Workshop on Learning with Natural Language Supervision, Proceedings of the Workshop
EditorsJacob Andreas, Karthik Narasimhan, Aida Nematzadeh
PublisherAssociation for Computational Linguistics (ACL)
Pages10-16
Number of pages7
ISBN (Electronic)9781955917452
StatePublished - 2022
Event1st Workshop on Learning with Natural Language Supervision, LNLS 2022 - Dublin, Ireland
Duration: 26 May 2022 → …

Publication series

NameLNLS 2022 - 1st Workshop on Learning with Natural Language Supervision, Proceedings of the Workshop

Conference

Conference1st Workshop on Learning with Natural Language Supervision, LNLS 2022
Country/TerritoryIreland
CityDublin
Period26/05/22 → …

Fingerprint

Dive into the research topics of 'GrammarSHAP: An Efficient Model-Agnostic and Structure-Aware NLP Explainer'. Together they form a unique fingerprint.

Cite this