Structure-Aware Language Model Pretraining Improves Dense Retrieval on Structured Data.
Xinze LiZhenghao LiuChenyan XiongShi YuYu GuZhiyuan LiuGe YuPublished in: CoRR (2023)
Keyphrases
- structured data
- language model
- retrieval model
- document retrieval
- structured information
- test collection
- ad hoc information retrieval
- language modeling
- information retrieval
- semi structured
- query expansion
- language models for information retrieval
- n gram
- probabilistic model
- information extraction
- smoothing methods
- query terms
- structured queries
- relevance model
- document length
- query specific
- unstructured data
- semi structured data
- linked data
- keyword search
- keyword queries
- data sources
- xml documents
- information retrieval systems
- textual data
- vector space model
- ad hoc retrieval
- image retrieval
- web pages
- retrieval effectiveness
- relational databases
- question answering
- database
- relevance feedback