MuRAR: A Simple and Effective Multimodal Retrieval and Answer Refinement Framework for Multimodal Question Answering

TL;DR

MuRAR is a framework that improves multimodal question answering by retrieving relevant data and refining answers to produce coherent, useful, and readable multimodal responses, especially for enterprise and educational applications.

Contribution

Introduces MuRAR, a simple framework that enhances multimodal answers in QA systems through retrieval and refinement, addressing limitations of previous text-focused approaches.

Findings

Multimodal answers are more useful and readable than plain text answers.

MuRAR effectively integrates multimodal data for comprehensive responses.

Human evaluations favor MuRAR-generated answers over traditional methods.

Abstract

Recent advancements in retrieval-augmented generation (RAG) have demonstrated impressive performance in the question-answering (QA) task. However, most previous works predominantly focus on text-based answers. While some studies address multimodal data, they still fall short in generating comprehensive multimodal answers, particularly for explaining concepts or providing step-by-step tutorials on how to accomplish specific goals. This capability is especially valuable for applications such as enterprise chatbots and settings such as customer service and educational systems, where the answers are sourced from multimodal data. In this paper, we introduce a simple and effective framework named MuRAR (Multimodal Retrieval and Answer Refinement). MuRAR enhances text-based answers by retrieving relevant multimodal data and refining the responses to create coherent multimodal answers. This framework can be easily extended to support multimodal answers in enterprise chatbots with minimal modifications. Human evaluation results indicate that multimodal answers generated by MuRAR are more useful and readable compared to plain text answers.