Scalable attentive sentence-pair modeling via distilled sentence embedding

Oren Barkan, Noam Razin, Itzik Malkiel, Ori Katz, Avi Caciularu, Noam Koenigstein

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

25 Scopus citations

Abstract

Recent state-of-the-art natural language understanding mod-els, such as BERT and XLNet, score a pair of sentences (A and B) using multiple cross-attention operations - a process in which each word in sentence A attends to all words in sentence B and vice versa. As a result, computing the simi-larity between a query sentence and a set of candidate sen-tences, requires the propagation of all query-candidate sen-tence-pairs throughout a stack of cross-attention layers. This exhaustive process becomes computationally prohibitive when the number of candidate sentences is large. In con-trast, sentence embedding techniques learn a sentence-to-vector mapping and compute the similarity between the sen-tence vectors via simple elementary operations. In this pa-per, we introduce Distilled Sentence Embedding (DSE) - a model that is based on knowledge distillation from cross-attentive models, focusing on sentence-pair tasks. The out-line of DSE is as follows: Given a cross-attentive teacher model (e.g. a fine-tuned BERT), we train a sentence embed-ding based student model to reconstruct the sentence-pair scores obtained by the teacher model. We empirically demonstrate the effectiveness of DSE on five GLUE sen-tence-pair tasks. DSE significantly outperforms several ELMO variants and other sentence embedding methods, while accelerating computation of the query-candidate sen-tence-pairs similarities by several orders of magnitude, with an average relative degradation of 4.6% compared to BERT. Furthermore, we show that DSE produces sentence embed-dings that reach state-of-the-art performance on universal sentence representation benchmarks. Our code is made pub-licly available at https://github.com/microsoft/Distilled-Sentence-Embedding.

Original languageEnglish
Title of host publicationAAAI 2020 - 34th AAAI Conference on Artificial Intelligence
Pages3235-3242
Number of pages8
ISBN (Electronic)9781577358350
StatePublished - 2020
Externally publishedYes
Event34th AAAI Conference on Artificial Intelligence, AAAI 2020 - New York, United States
Duration: 7 Feb 202012 Feb 2020

Publication series

NameAAAI 2020 - 34th AAAI Conference on Artificial Intelligence

Conference

Conference34th AAAI Conference on Artificial Intelligence, AAAI 2020
Country/TerritoryUnited States
CityNew York
Period7/02/2012/02/20

Fingerprint

Dive into the research topics of 'Scalable attentive sentence-pair modeling via distilled sentence embedding'. Together they form a unique fingerprint.

Cite this