FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction.
Chen-Yu LeeChun-Liang LiHao ZhangTimothy DozatVincent PerotGuolong SuXiang ZhangKihyuk SohnNikolay GlushnevRenshen WangJoshua AinslieShangbang LongSiyang QinYasuhisa FujiiNan HuaTomas PfisterPublished in: ACL (1) (2023)
Keyphrases
- information extraction
- prior knowledge
- learning systems
- learning algorithm
- learning process
- precision and recall
- text documents
- relational learning
- machine learning
- document representation
- document classification
- web mining
- conditional random fields
- multi modal
- online learning
- graphical models
- reinforcement learning