MS-Transformer: Introduce multiple structural priors into a unified transformer for encoding sentences

Transformers have been widely utilized in recent NLP studies. Unlike CNNs or RNNs, the vanilla Transformer is position-insensitive, and thus is incapable of capturing the structural priors between sequences of words. Existing studies commonly apply one single mask strategy on Transformers for incorporating structural priors while failing at modeling more abundant structural information of texts. In this paper, we aim at introducing multiple types of structural priors into Transformers, proposing the Multiple Structural Priors Guided Transformer (MS-Transformer) that transforms different structural priors into different attention heads by using a novel multi-mask based multi-head attention mechanism. In particular, we integrate two categories of structural priors, including the sequential order and the relative position of words. For the purpose of capturing the latent hierarchical structure of the texts, we extract these information not only from the word contexts but also from the dependency syntax trees. Experimental results on three tasks show that MS-Transformer achieves significant improvements against other strong baselines.

Automated summary

This paper introduces a Multiple Structural Priors Guided Transformer (MS-Transformer) that integrates different types of structural priors into Transformers using a novel multi-head attention mechanism, aiming to capture the rich structural information of texts. Experimental results demonstrate significant performance improvements of MS-Transformer compared to other models.