Noisy Self-Training with Synthetic Queries for Dense Retrieval.
Fan JiangTom DrummondTrevor CohnPublished in: EMNLP (Findings) (2023)
Keyphrases
- retrieval systems
- query formulation
- retrieval process
- retrieval quality
- query generation
- web retrieval
- boolean queries
- semantic query
- information retrieval
- data retrieval
- information retrieval systems
- ranked retrieval
- retrieval strategies
- trec web
- multiple queries
- response time
- exact match
- query processing
- original query
- content and structure
- user queries
- image database
- retrieval model
- query performance prediction
- query language
- query terms
- ad hoc retrieval
- indexing techniques
- database
- noisy observations
- indexing structure
- text retrieval
- test collection
- web search engines
- image retrieval
- query expansion
- query logs
- document retrieval
- range queries
- cross language retrieval
- training set
- retrieve documents
- structured documents
- semi supervised
- query specific
- documents retrieved
- data sources
- retrieval method
- search interface
- xml retrieval
- retrieved images
- query evaluation
- keyword queries
- language model
- approximate matches
- relevance feedback