Skip to Main content Skip to Navigation
Conference papers

Do Multilingual Neural Machine Translation Models Contain Language Pair Specific Attention Heads?

Abstract : Recent studies on the analysis of the multilingual representations focus on identifying whether there is an emergence of languageindependent representations, or whether a multilingual model partitions its weights among different languages. While most of such work has been conducted in a "black-box" manner, this paper aims to analyze individual components of a multilingual neural translation (NMT) model. In particular, we look at the encoder self-attention and encoder-decoder attention heads (in a many-to-one NMT model) that are more specific to the translation of a certain language pair than others by (1) employing metrics that quantify some aspects of the attention weights such as "variance" or "confidence", and (2) systematically ranking the importance of attention heads with respect to translation quality. Experimental results show that surprisingly, the set of most important attention heads are very similar across the language pairs and that it is possible to remove nearly one-third of the less important heads without hurting the translation quality greatly.
Document type :
Conference papers
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03299010
Contributor : Didier Schwab <>
Submitted on : Sunday, July 25, 2021 - 7:15:10 PM
Last modification on : Tuesday, July 27, 2021 - 3:38:41 AM

File

Do_Multilingual_Neural_Machine...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03299010, version 1

Collections

UGA | CNRS | LIG

Citation

Zae Kim, Laurent Besacier, Vassilina Nikoulina, Didier Schwab. Do Multilingual Neural Machine Translation Models Contain Language Pair Specific Attention Heads?. Findings of ACL 2021, Aug 2021, Bangkok (virtual), Thailand. ⟨hal-03299010⟩

Share

Metrics

Record views

17

Files downloads

35