Selecting Large Language Model to Fine-tune via Rectified Scaling Law.
Haowei LinBaizhou HuangHaotian YeQinyu ChenZihao WangSujian LiJianzhu MaXiaojun WanJames ZouYitao LiangPublished in: CoRR (2024)
Keyphrases
- language model
- fine tune
- language modeling
- fine tuning
- probabilistic model
- document retrieval
- n gram
- information retrieval
- language modelling
- query expansion
- speech recognition
- mixture model
- ad hoc information retrieval
- smoothing methods
- statistical language models
- test collection
- translation model
- language model for information retrieval
- context sensitive
- retrieval model
- pseudo relevance feedback
- vector space model
- query specific
- dirichlet prior
- document length
- bayesian networks
- pid controller
- query terms
- generative model
- text classification
- language models for information retrieval