DropBP: Accelerating Fine-Tuning of Large Language Models by Dropping Backward Propagation.
Sunghyeon WooBaeseong ParkByeongwook KimMinjung JoSejung KwonDongsuk JeonDongsoo LeePublished in: CoRR (2024)
Keyphrases
- fine tuning
- language model
- language modeling
- probabilistic model
- viable alternative
- n gram
- document retrieval
- speech recognition
- information retrieval
- fine tune
- language modelling
- query expansion
- retrieval model
- statistical language models
- document ranking
- context sensitive
- fine tuned
- language model for information retrieval
- ad hoc information retrieval
- test collection
- query terms
- domain specific
- vector space model
- translation model
- query specific
- smoothing methods
- word error rate
- pseudo relevance feedback
- term dependencies
- okapi bm
- statistical language modeling
- relevance model