FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction.
Chen-Yu LeeChun-Liang LiHao ZhangTimothy DozatVincent PerotGuolong SuXiang ZhangKihyuk SohnNikolai GlushnevRenshen WangJoshua AinslieShangbang LongSiyang QinYasuhisa FujiiNan HuaTomas PfisterPublished in: CoRR (2023)
Keyphrases
- information extraction
- learning algorithm
- learning process
- text summarization
- online learning
- document images
- learning systems
- supervised learning
- relational learning
- question answering
- random walk
- natural language processing
- digital libraries
- reinforcement learning
- text documents
- precision and recall
- free text
- graph model
- search engine
- machine learning