Constructing Multilingual Visual-Text Datasets Revealing Visual Multilingual Ability of Vision Language Models

TL;DR

This paper develops multilingual visual-text datasets across four languages to systematically evaluate vision-language models, introducing new tasks, rationales, and human evaluation to analyze their fine-grained visual linguistic abilities.

Contribution

It creates new multilingual datasets with rationales for comprehensive VLM evaluation, including the novel 'unrelatedness' task and analysis in Swahili and Urdu.

Findings

VLMs can be fine-tuned on the new datasets

Rationales improve understanding of VLM reasoning

First analysis of Swahili and Urdu VLM capabilities

Abstract

Large language models (LLMs) have increased interest in vision language models (VLMs), which process image-text pairs as input. Studies investigating the visual understanding ability of VLMs have been proposed, but such studies are still preliminary because existing datasets do not permit a comprehensive evaluation of the fine-grained visual linguistic abilities of VLMs across multiple languages. To further explore the strengths of VLMs, such as GPT-4V \cite{openai2023GPT4}, we developed new datasets for the systematic and qualitative analysis of VLMs. Our contribution is four-fold: 1) we introduced nine vision-and-language (VL) tasks (including object recognition, image-text matching, and more) and constructed multilingual visual-text datasets in four languages: English, Japanese, Swahili, and Urdu through utilizing templates containing \textit{questions} and prompting GPT4-V to generate the \textit{answers} and the \textit{rationales}, 2) introduced a new VL task named \textit{unrelatedness}, 3) introduced rationales to enable human understanding of the VLM reasoning process, and 4) employed human evaluation to measure the suitability of proposed datasets for VL tasks. We show that VLMs can be fine-tuned on our datasets. Our work is the first to conduct such analyses in Swahili and Urdu. Also, it introduces \textit{rationales} in VL analysis, which played a vital role in the evaluation.