PLaD: Preference-based Large Language Model Distillation with Pseudo-Preference Pairs.
Rongzhi ZhangJiaming ShenTianqi LiuHaorui WangZhen QinFeng HanJialu LiuSimon BaumgartnerMichael BenderskyChao ZhangPublished in: ACL (Findings) (2024)
Keyphrases
- language model
- language modeling
- user preferences
- preference relations
- n gram
- probabilistic model
- document retrieval
- query expansion
- speech recognition
- language modelling
- test collection
- mixture model
- information retrieval
- pairwise
- retrieval model
- statistical language models
- vector space model
- smoothing methods
- context sensitive
- language model for information retrieval
- query terms
- document ranking
- pseudo relevance feedback
- word error rate
- ad hoc information retrieval
- language models for information retrieval
- pseudo feedback
- statistical model
- feature selection