An Interactive Multi-modal Query Answering System with Retrieval-Augmented Large Language Models

Abstract

Retrieval-augmented Large Language Models (LLMs) have reshaped traditional query-answering systems, offering unparalleled user experiences. However, existing retrieval techniques often struggle to handle multi-modal query contexts. In this paper, we demonstrate an interactive Multi-modal Query Answering (MQA) system, empowered by our newly developed multi-modal retrieval framework and navigation graph index, and any cutting-edge LLMs. The MQA system comprises five core components—Data Preprocessing, Vector Representation, Index Construction, Query Execution, and Answer Generation, all orchestrated by a dedicated coordinator to ensure smooth data flow from input to answer generation. One notable aspect of MQA is its utilization of contrastive learning to assess the significance of different modalities, facilitating precise measurement of multi-modal information similarity. Furthermore, the system achieves efficient retrieval through our advanced navigation graph index, refined using computational pruning techniques. An additional highlight of our system is its pluggable processing framework, allowing seamless integration of embedding models, graph indexes, and LLMs. This flexibility provides users with diverse options for gaining insights from their multi-modal knowledge base. A preliminary video introduction of MQA is available at https://youtu.be/xvUuo2ZIqWk.

Mengzhao Wang

PhD candidate

I am currently a second-year Ph.D. student at Zhejiang University. My research interests include high-dimensional data storage, retrieval, and their applications in large language models (LLMs).