UniRAG: Universal Retrieval Augmentation for Large Vision Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sharifymoghaddam, Sahel, Upadhyay, Shivani, Chen, Wenhu, Lin, Jimmy
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929749702475776
author Sharifymoghaddam, Sahel
Upadhyay, Shivani
Chen, Wenhu
Lin, Jimmy
author_facet Sharifymoghaddam, Sahel
Upadhyay, Shivani
Chen, Wenhu
Lin, Jimmy
contents Recently, Large Vision Language Models (LVLMs) have unlocked many complex use cases that require Multi-Modal (MM) understanding (e.g., image captioning or visual question answering) and MM generation (e.g., text-guided image generation or editing) capabilities. To further improve the output fidelityof LVLMs we introduce UniRAG, a plug-and-play technique that adds relevant retrieved information to prompts as few-shot examples during inference. Unlike the common belief that Retrieval Augmentation (RA) mainly improves generation or understanding of uncommon entities, our evaluation results on the MSCOCO dataset with common entities show that both proprietary models like GPT-4o and Gemini-Pro and smaller open-source models like LLaVA, LaVIT, and Emu2 significantly enhance their generation quality when their input prompts are augmented with relevant information retrieved by Vision-Language (VL) retrievers like UniIR models. All the necessary code to reproduce our results is available at https://github.com/castorini/UniRAG
format Preprint
id arxiv_https___arxiv_org_abs_2405_10311
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle UniRAG: Universal Retrieval Augmentation for Large Vision Language Models
Sharifymoghaddam, Sahel
Upadhyay, Shivani
Chen, Wenhu
Lin, Jimmy
Information Retrieval
Recently, Large Vision Language Models (LVLMs) have unlocked many complex use cases that require Multi-Modal (MM) understanding (e.g., image captioning or visual question answering) and MM generation (e.g., text-guided image generation or editing) capabilities. To further improve the output fidelityof LVLMs we introduce UniRAG, a plug-and-play technique that adds relevant retrieved information to prompts as few-shot examples during inference. Unlike the common belief that Retrieval Augmentation (RA) mainly improves generation or understanding of uncommon entities, our evaluation results on the MSCOCO dataset with common entities show that both proprietary models like GPT-4o and Gemini-Pro and smaller open-source models like LLaVA, LaVIT, and Emu2 significantly enhance their generation quality when their input prompts are augmented with relevant information retrieved by Vision-Language (VL) retrievers like UniIR models. All the necessary code to reproduce our results is available at https://github.com/castorini/UniRAG
title UniRAG: Universal Retrieval Augmentation for Large Vision Language Models
topic Information Retrieval
url https://arxiv.org/abs/2405.10311