Seeing Through the MiRAGE: Evaluating Multimodal Retrieval Augmented Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Martin, Alexander, Walden, William, Kriz, Reno, Zhang, Dengjia, Sanders, Kate, Yang, Eugene, Jin, Chihsheng, Van Durme, Benjamin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911740306915328
author Martin, Alexander
Walden, William
Kriz, Reno
Zhang, Dengjia
Sanders, Kate
Yang, Eugene
Jin, Chihsheng
Van Durme, Benjamin
author_facet Martin, Alexander
Walden, William
Kriz, Reno
Zhang, Dengjia
Sanders, Kate
Yang, Eugene
Jin, Chihsheng
Van Durme, Benjamin
contents We introduce MiRAGE, an evaluation framework for retrieval-augmented generation (RAG) from multimodal sources. As audiovisual media becomes a prevalent source of information online, it is essential for RAG systems to integrate information from these sources into generation. However, existing evaluations for RAG are text-centric, limiting their applicability to multimodal settings. MiRAGE is a claim-centric approach to multimodal RAG evaluation, consisting of InfoF1, which assesses factuality and information coverage, and CiteF1, which assesses citation support and completeness. We show that, when applied by humans, MiRAGE strongly aligns with extrinsic judgments of output quality. We additionally introduce an automatic implementation of MiRAGE as well as multimodal variants of three prominent text-based RAG metrics -- ALCE, ARGUE, and RAGAS -- demonstrating the limitations of text-centric work and laying the groundwork for automatic evaluation. We release open-source implementations and outline evaluation methods for multimodal RAG.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24870
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Seeing Through the MiRAGE: Evaluating Multimodal Retrieval Augmented Generation
Martin, Alexander
Walden, William
Kriz, Reno
Zhang, Dengjia
Sanders, Kate
Yang, Eugene
Jin, Chihsheng
Van Durme, Benjamin
Computation and Language
Computer Vision and Pattern Recognition
Information Retrieval
We introduce MiRAGE, an evaluation framework for retrieval-augmented generation (RAG) from multimodal sources. As audiovisual media becomes a prevalent source of information online, it is essential for RAG systems to integrate information from these sources into generation. However, existing evaluations for RAG are text-centric, limiting their applicability to multimodal settings. MiRAGE is a claim-centric approach to multimodal RAG evaluation, consisting of InfoF1, which assesses factuality and information coverage, and CiteF1, which assesses citation support and completeness. We show that, when applied by humans, MiRAGE strongly aligns with extrinsic judgments of output quality. We additionally introduce an automatic implementation of MiRAGE as well as multimodal variants of three prominent text-based RAG metrics -- ALCE, ARGUE, and RAGAS -- demonstrating the limitations of text-centric work and laying the groundwork for automatic evaluation. We release open-source implementations and outline evaluation methods for multimodal RAG.
title Seeing Through the MiRAGE: Evaluating Multimodal Retrieval Augmented Generation
topic Computation and Language
Computer Vision and Pattern Recognition
Information Retrieval
url https://arxiv.org/abs/2510.24870