Unifying Multimodal Retrieval via Document Screenshot Embedding.
Xueguang MaSheng-Chieh LinMinghan LiWenhu ChenJimmy LinPublished in: CoRR (2024)
Keyphrases
- retrieval systems
- information retrieval
- document retrieval
- information retrieval systems
- structured documents
- multimedia documents
- document collections
- retrieval quality
- retrieval engine
- text retrieval
- document analysis
- index terms
- retrieval strategies
- image database
- document ranking
- document processing
- document images
- multi modal
- trec web
- effective retrieval
- document content
- document structure
- test collection
- document indexing
- query terms
- query specific
- content and structure
- music retrieval
- term weighting
- retrieval model
- passage retrieval
- relevant documents
- web documents
- image retrieval
- multimedia
- web retrieval
- trec genomics
- average precision
- tf idf
- document clustering
- query expansion
- digital libraries
- document level
- document representation
- content based retrieval
- term frequency
- handwritten documents
- audio visual
- vector space
- cf loadingtexthtml
- language model