A Bounding Box is Worth One Token: Interleaving Layout and Text in a Large Language Model for Document Understanding.
Jinghui LuHaiyang YuYanjie WangYongjie YeJingqun TangZiwei YangBinghong WuQi LiuHao FengHan WangHao LiuCan HuangPublished in: CoRR (2024)
Keyphrases
- language model
- bounding box
- information retrieval
- automatic text summarization
- language modeling
- document retrieval
- automatic summarization
- n gram
- query expansion
- probabilistic model
- retrieval model
- text retrieval
- test collection
- text mining
- relevance model
- query terms
- object categories
- vector space model
- document representation
- keywords
- information retrieval systems
- language independent
- object recognition
- relevant documents
- object classes