Progressively Optimized Bi-Granular Document Representation for Scalable Embedding Based Retrieval.
Shitao XiaoZheng LiuWeihao HanJianjin ZhangChaozhuo LiYingxia ShaoDefu LianXing XieHao SunDenvy DengLiangjie ZhangQi ZhangPublished in: CoRR (2022)
Keyphrases
- document representation
- vector space
- index terms
- document content
- retrieval model
- bag of words
- vector space model
- document collections
- language model
- document clustering
- information retrieval
- text documents
- information retrieval systems
- data fusion
- web documents
- test collection
- query expansion
- semantic information
- text classification
- relevance ranking
- relevance feedback
- text data
- retrieval systems
- background knowledge
- language modeling
- semantic relations
- similarity search
- image retrieval