Sorted LLaMA: Unlocking the Potential of Intermediate Layers of Large Language Models for Dynamic Inference Using Sorted Fine-Tuning (SoFT).
Parsa KavehzadehMojtaba ValipourMarzieh TahaeiAli GhodsiBoxing ChenMehdi RezagholizadehPublished in: CoRR (2023)
Keyphrases
- language model
- fine tuning
- language modeling
- probabilistic model
- test collection
- language modelling
- viable alternative
- n gram
- speech recognition
- retrieval model
- document retrieval
- language models for information retrieval
- query expansion
- information retrieval
- bayesian networks
- learning perl
- context sensitive
- statistical language models
- document ranking
- smoothing methods
- relevance model
- ad hoc information retrieval
- co occurrence
- spoken term detection