Visual Grounding of Inter-lingual Word-Embeddings

TL;DR

This paper explores how visual grounding can enhance multilingual word embeddings by aligning visual and textual data across English, Arabic, and German, revealing mixed effects on different language pairs and benchmark tasks.

Contribution

It introduces an implicit alignment method for inter-lingual visual grounding of word embeddings and evaluates its impact across three languages on various benchmarks.

Findings

Inter-lingual grounding improves embeddings for similar languages like English and German.

Grounding with Arabic sometimes degrades performance on word similarity tasks.

Arabic shows significant improvement in categorization benchmarks when grounded with English.

Abstract

Visual grounding of Language aims at enriching textual representations of language with multiple sources of visual knowledge such as images and videos. Although visual grounding is an area of intense research, inter-lingual aspects of visual grounding have not received much attention. The present study investigates the inter-lingual visual grounding of word embeddings. We propose an implicit alignment technique between the two spaces of vision and language in which inter-lingual textual information interacts in order to enrich pre-trained textual word embeddings. We focus on three languages in our experiments, namely, English, Arabic, and German. We obtained visually grounded vector representations for these languages and studied whether visual grounding on one or multiple languages improved the performance of embeddings on word similarity and categorization benchmarks. Our experiments suggest that inter-lingual knowledge improves the performance of grounded embeddings in similar languages such as German and English. However, inter-lingual grounding of German or English with Arabic led to a slight degradation in performance on word similarity benchmarks. On the other hand, we observed an opposite trend on categorization benchmarks where Arabic had the most improvement on English. In the discussion section, several reasons for those findings are laid out. We hope that our experiments provide a baseline for further research on inter-lingual visual grounding.