ConfigILM: A general purpose configurable library for combining image and language models for visual question answering.
Leonard W. HackelKai Norman ClasenBegüm DemirPublished in: SoftwareX (2024)
Keyphrases
- visual information
- question answering
- language model
- passage retrieval
- visual data
- low level
- audio visual
- visual features
- information retrieval
- image classification
- sentence retrieval
- image retrieval
- language modeling
- probabilistic model
- image features
- image content
- natural language processing
- n gram
- named entities
- natural language
- document retrieval
- image representation
- speech recognition
- information extraction
- query expansion
- question classification
- retrieval model
- pseudo relevance feedback
- question answering systems
- cross language
- machine learning
- relevance model