DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving.
Yinmin ZhongShengyu LiuJunda ChenJianbo HuYibo ZhuXuanzhe LiuXin JinHao ZhangPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- n gram
- probabilistic model
- information retrieval
- speech recognition
- query expansion
- document retrieval
- retrieval model
- test collection
- language modelling
- mixture model
- ad hoc information retrieval
- context sensitive
- vector space model
- language model for information retrieval
- smoothing methods
- translation model
- query terms
- word error rate
- statistical language models
- document ranking
- document length
- dirichlet prior
- language models for information retrieval
- generative model