Recognizing Everything from All Modalities at Once: Grounded Multimodal Universal Information Extraction.
Meishan ZhangHao FeiBin WangShengqiong WuYixin CaoFei LiMin ZhangPublished in: ACL (Findings) (2024)
Keyphrases
- information extraction
- multiple modalities
- multimodal fusion
- multi modal
- cross modal
- single modality
- natural language processing
- multimodal biometrics
- precision and recall
- named entity recognition
- multimodal data
- multimodal interaction
- information retrieval
- semi structured
- structured data
- text mining
- natural language
- named entities
- question answering
- relation extraction
- machine learning
- free text
- web mining
- conditional random fields
- natural language text
- high robustness
- text documents
- ontology based information extraction
- semantic tagging
- web documents
- multimodal interfaces
- open domain
- text processing
- textual data
- data extraction
- text summarization
- graphical models
- relevance feedback
- hidden markov models
- learning algorithm
- audio visual
- brain image analysis
- word sense disambiguation