mlsanityData to 14 Oct 2026

KoViDoRe: A Benchmark for Korean Visual Document Retrieval

arXiv 2608.20840 ↗First posted 21 Aug 2026cs.IR cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Recent advances in multimodal retrieval have improved the ability to retrieve information from visually rich documents such as PDFs and reports. However, existing benchmarks remain largely centered on English and provide limited coverage of Korean visual documents with complex structures. Furthermore, most existing Korean resources primarily evaluate single-page retrieval, failing to capture realistic scenarios that require evidence aggregation across multiple pages. To address these gaps, we introduce KoViDoRe, a benchmark for Korean visual document retrieval. The dataset is constructed from publicly available Korean documents with diverse layouts, including tables, figures, and multi-column structures. We develop a multi-stage data curation pipeline consisting of structured document parsing, synthetic query generation using both summary-based and context-based strategies, and relevance mapping with human verification. Using KoViDoRe, we evaluate a wide range of multimodal retrieval models and observe that current models struggle to effectively handle Korean visual document retrieval, particularly in settings involving structured content and diverse query types. Motivated by this finding, we further curate a large-scale training dataset, Ko-VDR Train Public, to support the development of retrieval models tailored to Korean visual documents. Together, KoViDoRe and Ko-VDR Train Public provide a unified benchmark and training resource for Korean visual document retrieval.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 30 references that are in the map. First posted on arXiv 21 Aug 2026.

Builds on

20 of 30 references.

References · 20

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  3. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,252
  4. 2017FigureQA: An Annotated Figure Dataset for Visual Reasoningcited by 319
  5. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documentscited by 245
  6. 2026Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Rankingcited by 230
  7. 2024VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Taskscited by 228
  8. 2024M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understandingcited by 111
  9. 2024Unifying Multimodal Retrieval via Document Screenshot Embeddingcited by 110
  10. 2025jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrievalcited by 76
  11. 2024TableVQA-Bench: A Visual Question Answering Benchmark on Multiple Table Domainscited by 69
  12. 2025ViDoRAG: Visual Document Retrieval-Augmented Generation via Dynamic Iterative Reasoning Agentscited by 66
8 more
  1. 2025ViDoRe Benchmark V2: Raising the Bar for Visual Retrievalcited by 61
  2. 2025Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generationcited by 42
  3. 2024jina-clip-v2: Multilingual Multimodal Embeddings for Text and Imagescited by 37
  4. 2025REAL-MM-RAG: A Real-World Multi-Modal Retrieval Benchmarkcited by 35
  5. 2025UNIDOC-BENCH: A Unified Benchmark for Document-Centric Multimodal RAGcited by 21
  6. 2025MIRACL-VISION: A Large, multilingual, visual document retrieval benchmarkcited by 9
  7. 2026Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrievalcited by 8
  8. 2025SDS KoPub VDR: A Benchmark Dataset for Visual Document Retrieval in Korean Public Documentscited by 4

Built on by

Not cited yet.