OCRBench

What is OCRBench?

OCRBench provides a comprehensive evaluation of Large Multimodal Models, such as GPT4V and Gemini, in various text-related visual tasks including Text…

Released

2024-01-17

Evaluates

语言, Language, 理解, Understanding, 多模态模型, VLM, 语言理解, Comprehension, 图像理解, Image Understanding, OCR与文档理解, OCR and Document Understanding, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/OCRBench

OCRBench paper and code

• Paper https://arxiv.org/abs/2305.07895
• Code https://github.com/Yuliang-Liu/MultimodalOCR

Which sources cite OCRBench?

• OCRBench

Related 语言, Language, 理解, Understanding, 多模态模型, VLM, 语言理解, Comprehension, 图像理解, Image Understanding, OCR与文档理解, OCR and Document Understanding, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard