Fine-Grained Human Feedback Gives Better Rewards for Language Model Training.
Zeqiu WuYushi HuWeijia ShiNouha DziriAlane SuhrPrithviraj AmmanabroluNoah A. SmithMari OstendorfHannaneh HajishirziPublished in: CoRR (2023)
Keyphrases
- fine grained
- language model
- language modeling
- coarse grained
- n gram
- document retrieval
- probabilistic model
- access control
- query expansion
- test collection
- speech recognition
- language modelling
- retrieval model
- information retrieval
- mixture model
- smoothing methods
- ad hoc information retrieval
- document ranking
- vector space model
- query terms
- context sensitive
- pseudo relevance feedback
- data lineage
- translation model
- relevance model
- user feedback
- retrieval effectiveness
- ranking algorithm
- relevance feedback
- bayesian networks
- high level