Semantics-aware Attention Improves Neural Machine Translation

Abstract

The integration of syntactic structures into Transformer machine translation has shown positive results, but to our knowledge, no work has attempted to do so with semantic structures. In this work we propose two novel parameter-free methods for injecting semantic information into Transformers, both rely on semantics-aware masking of (some of) the attention heads. One such method operates on the encoder, through a Scene-Aware Self-Attention (SASA) head. Another on the decoder, through a Scene-Aware Cross-Attention (SACrA) head. We show a consistent improvement over the vanilla Transformer and syntax-aware models for four language pairs. We further show an additional gain when using both semantic and syntactic structures in some language pairs.

arXiv commentAccepted to *SEM 2022

References in corpus (6)

• A Transition-Based Directed Acyclic Graph Parser for UCCA
• Semantic Neural Machine Translation using AMR
• Syntactically Supervised Transformers for Faster Neural Machine Translation
• Building Subject-aligned Comparable Corpora and Mining it for Truly Parallel Sentence Pairs
• Syntax-Infused Transformer and BERT models for Machine Translation and Natural Language Understanding
• Mediators in Determining what Processing BERT Performs First