Adapting Open-Source Large Language Models for Cost-Effective, Expert-Level Clinical Note Generation with On-Policy Reinforcement Learning.
Hanyin WangChufan GaoBolun LiuQiping XuGuleid HusseinMohamad El LabbanKingsley IheasirimHariprasad KorsapatiChuck OutcaltJimeng SunPublished in: CoRR (2024)
Keyphrases
- cost effective
- language model
- open source
- reinforcement learning
- language modeling
- optimal policy
- n gram
- document level
- cost effectiveness
- document retrieval
- speech recognition
- retrieval model
- low cost
- language modelling
- probabilistic model
- statistical language models
- query expansion
- information retrieval
- context sensitive
- test collection
- policy search
- pseudo relevance feedback
- inverse reinforcement learning
- language models for information retrieval
- document ranking
- ad hoc information retrieval
- smoothing methods
- cross lingual
- translation model
- query specific
- state space
- learning algorithm
- term dependencies
- spoken term detection
- document length
- real time
- markov decision processes
- retrieval effectiveness