Preference-grounded Token-level Guidance for Language Model Fine-tuning.
Shentao YangShujian ZhangCongying XiaYihao FengCaiming XiongMingyuan ZhouPublished in: CoRR (2023)
Keyphrases
- language model
- fine tuning
- language modeling
- n gram
- document retrieval
- speech recognition
- document level
- probabilistic model
- language modelling
- information retrieval
- query expansion
- statistical language models
- retrieval model
- test collection
- context sensitive
- language model for information retrieval
- pseudo relevance feedback
- query terms
- mixture model
- ad hoc information retrieval
- fine tuned
- relevance model
- smoothing methods
- pseudo feedback
- model selection
- vector space model
- query specific
- word error rate