Noisy Self-Training with Synthetic Queries for Dense Retrieval.
Fan JiangTom DrummondTrevor CohnPublished in: CoRR (2023)
Keyphrases
- retrieval systems
- query formulation
- retrieval process
- retrieval quality
- query generation
- web retrieval
- boolean queries
- multiple queries
- query language
- query evaluation
- information retrieval
- data retrieval
- indexing techniques
- original query
- trec web
- image database
- ranked retrieval
- user queries
- information retrieval systems
- retrieval strategies
- response time
- semi supervised learning
- database
- query specific
- query processing
- keyword queries
- query logs
- test collection
- database queries
- retrieval method
- content and structure
- ad hoc retrieval
- query expansion
- cross language retrieval
- documents retrieved
- training set
- image retrieval
- query refinement
- query terms
- range queries
- multimedia databases
- cost sensitive
- web search engines
- query performance prediction
- text queries
- semantic query
- exact match
- average precision
- handwritten documents
- document retrieval
- data sources
- web search
- learning algorithm
- retrieval effectiveness
- relevant documents
- relevance feedback
- noisy observations
- xml elements
- retrieval model