MIRACL-VISION: A Large, multilingual, visual document retrieval benchmark

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Osmulski, Radek, Moreira, Gabriel de Souza P., Ak, Ronay, Xu, Mengyao, Schifferer, Benedikt, Oldridge, Even
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909618609848320
author Osmulski, Radek
Moreira, Gabriel de Souza P.
Ak, Ronay
Xu, Mengyao
Schifferer, Benedikt
Oldridge, Even
author_facet Osmulski, Radek
Moreira, Gabriel de Souza P.
Ak, Ronay
Xu, Mengyao
Schifferer, Benedikt
Oldridge, Even
contents Document retrieval is an important task for search and Retrieval-Augmented Generation (RAG) applications. Large Language Models (LLMs) have contributed to improving the accuracy of text-based document retrieval. However, documents with complex layout and visual elements like tables, charts and infographics are not perfectly represented in textual format. Recently, image-based document retrieval pipelines have become popular, which use visual large language models (VLMs) to retrieve relevant page images given a query. Current evaluation benchmarks on visual document retrieval are limited, as they primarily focus only English language, rely on synthetically generated questions and offer a small corpus size. Therefore, we introduce MIRACL-VISION, a multilingual visual document retrieval evaluation benchmark. MIRACL-VISION covers 18 languages, and is an extension of the MIRACL dataset, a popular benchmark to evaluate text-based multilingual retrieval pipelines. MIRACL was built using a human-intensive annotation process to generate high-quality questions. In order to reduce MIRACL-VISION corpus size to make evaluation more compute friendly while keeping the datasets challenging, we have designed a method for eliminating the "easy" negatives from the corpus. We conducted extensive experiments comparing MIRACL-VISION with other benchmarks, using popular public text and image models. We observe a gap in state-of-the-art VLM-based embedding models on multilingual capabilities, with up to 59.7% lower retrieval accuracy than a text-based retrieval models. Even for the English language, the visual models retrieval accuracy is 12.1% lower compared to text-based models. MIRACL-VISION is a challenging, representative, multilingual evaluation benchmark for visual retrieval pipelines and will help the community build robust models for document retrieval.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11651
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MIRACL-VISION: A Large, multilingual, visual document retrieval benchmark
Osmulski, Radek
Moreira, Gabriel de Souza P.
Ak, Ronay
Xu, Mengyao
Schifferer, Benedikt
Oldridge, Even
Information Retrieval
Computer Vision and Pattern Recognition
Document retrieval is an important task for search and Retrieval-Augmented Generation (RAG) applications. Large Language Models (LLMs) have contributed to improving the accuracy of text-based document retrieval. However, documents with complex layout and visual elements like tables, charts and infographics are not perfectly represented in textual format. Recently, image-based document retrieval pipelines have become popular, which use visual large language models (VLMs) to retrieve relevant page images given a query. Current evaluation benchmarks on visual document retrieval are limited, as they primarily focus only English language, rely on synthetically generated questions and offer a small corpus size. Therefore, we introduce MIRACL-VISION, a multilingual visual document retrieval evaluation benchmark. MIRACL-VISION covers 18 languages, and is an extension of the MIRACL dataset, a popular benchmark to evaluate text-based multilingual retrieval pipelines. MIRACL was built using a human-intensive annotation process to generate high-quality questions. In order to reduce MIRACL-VISION corpus size to make evaluation more compute friendly while keeping the datasets challenging, we have designed a method for eliminating the "easy" negatives from the corpus. We conducted extensive experiments comparing MIRACL-VISION with other benchmarks, using popular public text and image models. We observe a gap in state-of-the-art VLM-based embedding models on multilingual capabilities, with up to 59.7% lower retrieval accuracy than a text-based retrieval models. Even for the English language, the visual models retrieval accuracy is 12.1% lower compared to text-based models. MIRACL-VISION is a challenging, representative, multilingual evaluation benchmark for visual retrieval pipelines and will help the community build robust models for document retrieval.
title MIRACL-VISION: A Large, multilingual, visual document retrieval benchmark
topic Information Retrieval
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.11651