MLM: A Benchmark Dataset for Multitask Learning with Multiple Languages and Modalities

TL;DR

This paper introduces MLM, a comprehensive multilingual and multimodal dataset designed to evaluate and advance multitask learning systems across diverse languages and data types, with applications in digital humanities and beyond.

Contribution

The paper presents MLM, a novel benchmark dataset that includes multiple languages and modalities, along with a geo-representative subset, to facilitate research in multitask learning and multimodal understanding.

Findings

Baseline systems struggle to generalize across diverse data.

The dataset enables testing of multilingual and multimodal multitask models.

The geo-representative subset aids location-based research.

Abstract

In this paper, we introduce the MLM (Multiple Languages and Modalities) dataset - a new resource to train and evaluate multitask systems on samples in multiple modalities and three languages. The generation process and inclusion of semantic data provide a resource that further tests the ability for multitask systems to learn relationships between entities. The dataset is designed for researchers and developers who build applications that perform multiple tasks on data encountered on the web and in digital archives. A second version of MLM provides a geo-representative subset of the data with weighted samples for countries of the European Union. We demonstrate the value of the resource in developing novel applications in the digital humanities with a motivating use case and specify a benchmark set of tasks to retrieve modalities and locate entities in the dataset. Evaluation of baseline multitask and single task systems on the full and geo-representative versions of MLM demonstrate the challenges of generalising on diverse data. In addition to the digital humanities, we expect the resource to contribute to research in multimodal representation learning, location estimation, and scene understanding.