UDA: A Benchmark Suite for Retrieval Augmented Generation in Real-world Document Analysis.
Yulong HuiYao LuHuanchen ZhangPublished in: CoRR (2024)
Keyphrases
- document analysis
- benchmark suite
- real world
- document images
- document image analysis
- image analysis
- text analysis
- character recognition
- document processing
- word recognition
- document image retrieval
- electronic documents
- printed documents
- handwritten documents
- word segmentation
- word level
- search engine
- image classification
- pattern recognition
- image segmentation
- metadata
- data mining