Translation Deserves Better: Analyzing Translation Artifacts in Cross-lingual Visual Question Answering

TL;DR

This paper investigates how translation artifacts in machine-translated evaluation samples impact cross-lingual visual question answering systems and proposes a data augmentation method to mitigate these effects.

Contribution

It identifies the influence of translation artifacts on VQA models and introduces a simple augmentation technique to reduce their adverse impact.

Findings

Translation artifacts significantly affect model performance.

The proposed augmentation improves robustness across languages.

Artifacts differ from human-written text, influencing model behavior.

Abstract

Building a reliable visual question answering~(VQA) system across different languages is a challenging problem, primarily due to the lack of abundant samples for training. To address this challenge, recent studies have employed machine translation systems for the cross-lingual VQA task. This involves translating the evaluation samples into a source language (usually English) and using monolingual models (i.e., translate-test). However, our analysis reveals that translated texts contain unique characteristics distinct from human-written ones, referred to as translation artifacts. We find that these artifacts can significantly affect the models, confirmed by extensive experiments across diverse models, languages, and translation processes. In light of this, we present a simple data augmentation strategy that can alleviate the adverse impacts of translation artifacts.