MUST-VQA: MUltilingual Scene-text VQA

TL;DR

This paper introduces MUST-VQA, a multilingual scene text visual question answering framework capable of handling multiple languages in zero-shot scenarios, demonstrating competitive performance and effective adaptation of multilingual models.

Contribution

It proposes a generalized STVQA framework for multilingual, zero-shot language handling and evaluates its effectiveness across different scenarios.

Findings

Models perform well in zero-shot settings.

Multilingual models can be effectively adapted for STVQA.

The framework supports multiple languages without scene text alignment.

Abstract

In this paper, we present a framework for Multilingual Scene Text Visual Question Answering that deals with new languages in a zero-shot fashion. Specifically, we consider the task of Scene Text Visual Question Answering (STVQA) in which the question can be asked in different languages and it is not necessarily aligned to the scene text language. Thus, we first introduce a natural step towards a more generalized version of STVQA: MUST-VQA. Accounting for this, we discuss two evaluation scenarios in the constrained setting, namely IID and zero-shot and we demonstrate that the models can perform on a par on a zero-shot setting. We further provide extensive experimentation and show the effectiveness of adapting multilingual language models into STVQA tasks.