Large Language Models and Multimodal Retrieval for Visual Word Sense Disambiguation

TL;DR

This paper explores the use of multimodal transformers and Large Language Models to improve visual word sense disambiguation by combining various retrieval and reasoning approaches, leading to competitive results.

Contribution

It introduces a comprehensive approach combining multimodal retrieval, LLM-based knowledge enhancement, and learn-to-rank models for VWSD, advancing the state of the art.

Findings

Multimodal transformer methods improve retrieval accuracy.

LLMs with Chain-of-Thought enhance explainability.

Combining modules via learn-to-rank yields competitive performance.

Abstract

Visual Word Sense Disambiguation (VWSD) is a novel challenging task with the goal of retrieving an image among a set of candidates, which better represents the meaning of an ambiguous word within a given context. In this paper, we make a substantial step towards unveiling this interesting task by applying a varying set of approaches. Since VWSD is primarily a text-image retrieval task, we explore the latest transformer-based methods for multimodal retrieval. Additionally, we utilize Large Language Models (LLMs) as knowledge bases to enhance the given phrases and resolve ambiguity related to the target word. We also study VWSD as a unimodal problem by converting to text-to-text and image-to-image retrieval, as well as question-answering (QA), to fully explore the capabilities of relevant models. To tap into the implicit knowledge of LLMs, we experiment with Chain-of-Thought (CoT) prompting to guide explainable answer generation. On top of all, we train a learn to rank (LTR) model in order to combine our different modules, achieving competitive ranking results. Extensive experiments on VWSD demonstrate valuable insights to effectively drive future directions.