Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback.
Katherine TianEric MitchellAllan ZhouArchit SharmaRafael RafailovHuaxiu YaoChelsea FinnChristopher D. ManningPublished in: CoRR (2023)
Keyphrases
- language model
- fine tuned
- confidence scores
- language modeling
- n gram
- probabilistic model
- document retrieval
- retrieval model
- fine tuning
- language modelling
- query expansion
- information retrieval
- test collection
- context sensitive
- language models for information retrieval
- mixture model
- domain specific
- statistical language models
- smoothing methods
- relevance model
- graphical models
- spoken term detection