A Novel Graph-based Multi-modal Fusion Encoder for Neural Machine Translation

TL;DR

This paper introduces a graph-based multi-modal fusion encoder for neural machine translation that effectively captures semantic relationships between text and images, improving translation quality.

Contribution

It proposes a novel graph-based encoder that models fine-grained semantic correspondences between modalities for enhanced multi-modal NMT.

Findings

Outperforms existing multi-modal NMT models on Multi30K dataset.

Demonstrates improved translation accuracy through semantic interaction layers.

Provides in-depth analysis confirming the effectiveness of the graph-based approach.

Abstract

Multi-modal neural machine translation (NMT) aims to translate source sentences into a target language paired with images. However, dominant multi-modal NMT models do not fully exploit fine-grained semantic correspondences between semantic units of different modalities, which have potential to refine multi-modal representation learning. To deal with this issue, in this paper, we propose a novel graph-based multi-modal fusion encoder for NMT. Specifically, we first represent the input sentence and image using a unified multi-modal graph, which captures various semantic relationships between multi-modal semantic units (words and visual objects). We then stack multiple graph-based multi-modal fusion layers that iteratively perform semantic interactions to learn node representations. Finally, these representations provide an attention-based context vector for the decoder. We evaluate our proposed encoder on the Multi30K datasets. Experimental results and in-depth analysis show the superiority of our multi-modal NMT model.