KoViDoRe - VQA v1.0

This dataset is part of the KoViDoRe Benchmark for Korean visual document retrieval evaluation. Specifically, it is focused on Visual Question Answering (VQA) tasks for Korean document images.

  Dataset Summary

Domain: Korean structured document images (e.g., 환경기상, 공공행정 등 다양한 카테고리) Task: Visual document retrieval (text question → relevant document image) Format: BEIR-compatible (corpus, queries, qrels)

Statistics:

Number of Documents (Pages): 1,100+… See the full description on the dataset page: https://huggingface.co/datasets/whybe-choi/kovidore-vqa-v1.0-beir.

OCRBench is a comprehensive evaluation benchmark designed to assess the OCR capabilities of Large Multimodal Models. It comprises five components: Text Recognition, SceneText-Centric VQA, Document-Oriented VQA, Key Information Extraction, and Handwritten Mathematical Expression Recognition. The benchmark includes 1000 question-answer pairs, and all the answers undergo manual verification and correction to ensure a more precise evaluation.