Optimizing Speculative Decoding for Serving Large Language Models Using Goodput.
Xiaoxuan LiuCade DanielLangxiang HuWoosuk KwonZhuohan LiXiangxi MoAlvin CheungZhijie DengIon StoicaHao ZhangPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- n gram
- document retrieval
- context sensitive
- probabilistic model
- speech recognition
- query expansion
- retrieval model
- information retrieval
- query terms
- statistical language models
- ad hoc information retrieval
- test collection
- language models for information retrieval
- language modelling
- pseudo relevance feedback
- smoothing methods
- document length
- pseudo feedback
- error rate
- search engine