Tackling Ambiguity with Images: Improved Multimodal Machine Translation and Contrastive Evaluation

TL;DR

This paper introduces a new multimodal machine translation approach that leverages images and novel training techniques to better resolve ambiguity, achieving superior performance on a new contrastive evaluation dataset.

Contribution

The paper proposes a novel MMT model with neural adapters and guided self-attention, and introduces CoMMuTE, a dataset for evaluating disambiguation in multimodal translation.

Findings

Competitive results on standard benchmarks.

Significant improvements on the CoMMuTE contrastive test set.

Outperforms state-of-the-art MMT systems.

Abstract

One of the major challenges of machine translation (MT) is ambiguity, which can in some cases be resolved by accompanying context such as images. However, recent work in multimodal MT (MMT) has shown that obtaining improvements from images is challenging, limited not only by the difficulty of building effective cross-modal representations, but also by the lack of specific evaluation and training data. We present a new MMT approach based on a strong text-only MT model, which uses neural adapters, a novel guided self-attention mechanism and which is jointly trained on both visually-conditioned masking and MMT. We also introduce CoMMuTE, a Contrastive Multilingual Multimodal Translation Evaluation set of ambiguous sentences and their possible translations, accompanied by disambiguating images corresponding to each translation. Our approach obtains competitive results compared to strong text-only models on standard English-to-French, English-to-German and English-to-Czech benchmarks and outperforms baselines and state-of-the-art MMT systems by a large margin on our contrastive test set. Our code and CoMMuTE are freely available.