Explainability by Parsing: Neural Module Tree Networks for Natural Language Visual Grounding.
Daqing LiuHanwang ZhangZheng-Jun ZhaFeng WuPublished in: CoRR (2018)
Keyphrases
- natural language
- natural language processing
- neural network
- visual information
- knowledge representation
- natural language interface
- linguistic analysis
- visual processing
- tree structure
- machine learning
- information extraction
- natural language understanding
- parse tree
- network architecture
- recurrent networks
- spiking neural networks
- semantic analysis
- question answering
- visual features
- semantic representation
- low level
- spiking neurons
- natural language sentences
- language processing
- visual perception
- complex networks
- network structure
- syntactic structures
- highly ambiguous
- biologically motivated
- neural mechanisms
- primary visual cortex
- natural language parsing
- natural language generation
- learning rules
- bio inspired
- spanning tree
- index structure
- data structure