Is Crowdsourcing Breaking Your Bank? Cost-Effective Fine-Tuning of Pre-trained Language Models with Proximal Policy Optimization.
Shuo YangGjergji KasneciPublished in: LREC/COLING (2024)
Keyphrases
- cost effective
- fine tuning
- language model
- pre trained
- language modeling
- low cost
- document retrieval
- n gram
- probabilistic model
- speech recognition
- query expansion
- cost effectiveness
- information retrieval
- retrieval model
- language modelling
- smoothing methods
- statistical language models
- fine tuned
- test collection
- document ranking
- language models for information retrieval
- training examples
- training data
- relevance model
- statistical model
- image classification
- feature vectors
- pairwise
- computer vision
- data sets