Document-as-Image Representations Fall Short for Scientific Retrieval

• URL: http://arxiv.org/abs/2604.18508v1
• Date: Mon, 20 Apr 2026 17:00:17 GMT
• Title: Document-as-Image Representations Fall Short for Scientific Retrieval
• Abstract summary: We introduce ArXivDoc, a new benchmark constructed from the underlying sources of scientific papers.<n>We show that document-as-image representations are consistently suboptimal, especially as document length increases.<n>Text-based representations are most effective, even for figure-based queries, by leveraging captions and surrounding context.
• Score: 54.7382379182364
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many recent document embedding models are trained on document-as-image representations, embedding rendered pages as images rather than the underlying source. Meanwhile, existing benchmarks for scientific document retrieval, such as ArXivQA and ViDoRe, treat documents as images of pages, implicitly favoring such representations. In this work, we argue that this paradigm is not well-suited for text-rich multimodal scientific documents, where critical evidence is distributed across structured sources, including text, tables, and figures. To study this setting, we introduce ArXivDoc, a new benchmark constructed from the underlying LaTeX sources of scientific papers. Unlike PDF or image-based representations, LaTeX provides direct access to structured elements (e.g., sections, tables, figures, equations), enabling controlled query construction grounded in specific evidence types. We systematically compare text-only, image-based, and multimodal representations across both single-vector and multi-vector retrieval models. Our results show that: (1) document-as-image representations are consistently suboptimal, especially as document length increases; (2) text-based representations are most effective, even for figure-based queries, by leveraging captions and surrounding context; and (3) interleaved text+image representations outperform document-as-image approaches without requiring specialized training.

Related papers

• MonkeyOCRv2: A Visual-Text Foundation Model for Document AI [71.06391669976786]
  MonkeyOCRv2 is a visual-text pretrained model for document AI.<n> MonkeyDoc v2 is the largest document-image pretraining corpus, comprising 113 million images spanning 17 languages.
  arXiv  Detail & Related papers  (2026-07-13T13:43:39Z)
• MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing [74.84107522458798]
  MPDocBench-Parse is a benchmark for multi-page document parsing in real-world applications.<n>It contains 433 manually annotated documents with 3,246 pages, covering 15 document types in English and Chinese, with diverse layout styles.
  arXiv  Detail & Related papers  (2026-05-21T07:36:41Z)
• Multimodal OCR: Parse Anything from Documents [72.69545534962234]
  dots.mocr treats visual elements such as charts, diagrams, tables, and icons as first-class parsing targets.<n>It reconstructs both text and graphics as structured outputs, enabling more faithful document reconstruction.<n>It supports end-to-end training over heterogeneous document elements.
  arXiv  Detail & Related papers  (2026-03-13T14:42:21Z)
• DISCO: Document Intelligence Suite for COmparative Evaluation [1.4425299138308667]
  Document intelligence requires accurate text extraction and reliable reasoning over document content.<n>We evaluate optical character recognition (OCR) pipelines and vision-language models (VLMs) separately on parsing and question answering across diverse document types.
  arXiv  Detail & Related papers  (2026-03-04T14:47:34Z)
• UNIDOC-BENCH: A Unified Benchmark for Document-Centric Multimodal RAG [82.84014669683863]
  Multimodal retrieval-augmented generation (MM-RAG) is a key approach for applying large language models to real-world knowledge bases.<n>UniDoc-Bench is the first large-scale, realistic benchmark for MM-RAG built from 70k real-world PDF pages.<n>Our experiments show that multimodal text-image fusion RAG systems consistently outperform both unimodal and jointly multimodal embedding-based retrieval.
  arXiv  Detail & Related papers  (2025-10-04T04:30:13Z)
• ColPali: Efficient Document Retrieval with Vision Language Models [15.369861972085136]
  We introduce the Visual Document Retrieval Benchmark ViDoRe, composed of various page-level retrieval tasks spanning multiple domains, languages, and practical settings.<n>The inherent complexity and performance shortcomings of modern systems motivate a new concept; doing document retrieval by directly embedding the images of the document pages.<n>We release ColPali, a Vision Language Model trained to produce high-quality multi-vector embeddings from images of document pages.
  arXiv  Detail & Related papers  (2024-06-27T15:45:29Z)
• DocumentCLIP: Linking Figures and Main Body Text in Reflowed Documents [18.080447065002392]
  We propose DocumentCLIP to enforce vision-language pretraining models to comprehend the interaction between images and longer text within documents. Our model is beneficial for the real-world multimodal document understanding like news article, magazines, product descriptions, which contain linguistically and visually richer content.
  arXiv  Detail & Related papers  (2023-06-09T23:51:11Z)
• Multi-View Document Representation Learning for Open-Domain Dense Retrieval [87.11836738011007]
  This paper proposes a multi-view document representation learning framework. It aims to produce multi-view embeddings to represent documents and enforce them to align with different queries. Experiments show our method outperforms recent works and achieves state-of-the-art results.
  arXiv  Detail & Related papers  (2022-03-16T03:36:38Z)
• Towards a Multi-modal, Multi-task Learning based Pre-training Framework for Document Representation Learning [5.109216329453963]
  We introduce Document Topic Modelling and Document Shuffle Prediction as novel pre-training tasks. We utilize the Longformer network architecture as the backbone to encode the multi-modal information from multi-page documents in an end-to-end fashion.
  arXiv  Detail & Related papers  (2020-09-30T05:39:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.