A Cross-lingual Comparison of Human and Model Relative Word Importance

Felix Morger, Stephanie Brandl, Lisa Beinborn, Nora Hollenstein

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningpeer review

16 Downloads (Pure)

Abstract

Relative word importance is a key metric for natural language processing. In this work, we compare human and model relative word importance to investigate if pretrained neural language models focus on the same words as humans cross-lingually. We perform an extensive study using several importance metrics (gradient-based saliency and attention-based) in monolingual and multilingual models, including eye-tracking corpora from four languages (German, Dutch, English, and Russian). We find that gradient-based saliency, first-layer attention, and attention flow correlate strongly with human eye-tracking data across all four languages. We further analyze the role of word length and word frequency in determining relative importance and find that it strongly correlates with length and frequency, however, the mechanisms behind these non-linear relations remain elusive. We obtain a cross-lingual approximation of the similarity between human and computational language processing and insights into the usability of several importance metrics.
OriginalsprogEngelsk
TitelProceedings of the 2022 CLASP Conference on (Dis)embodiment
Antal sider13
ForlagAssociation for Computational Linguistics (ACL)
Publikationsdato2022
Sider11-23
StatusUdgivet - 2022
Begivenhed2022 CLASP Conference on (Dis)embodiment - Gothenburg / Online, Sverige
Varighed: 15 sep. 202216 sep. 2022

Konference

Konference2022 CLASP Conference on (Dis)embodiment
Land/OmrådeSverige
ByGothenburg / Online
Periode15/09/202216/09/2022

Citationsformater