An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs.
Haihao ShenHengyu MengBo DongZhe WangOfir ZafrirYi DingYu LuoHanwen ChangQun GaoZiheng WangGuy BoudoukhMoshe WasserblatPublished in: CoRR (2023)
Keyphrases
- language model
- language modeling
- n gram
- statistical language models
- probabilistic model
- document retrieval
- retrieval model
- information retrieval
- speech recognition
- language modelling
- context sensitive
- relevance model
- smoothing methods
- mixture model
- language models for information retrieval
- query expansion
- query terms
- high dimensional
- parallel implementation
- document ranking
- ad hoc information retrieval
- vector space model
- query specific
- okapi bm
- test collection
- language model for information retrieval
- term dependencies
- translation model
- pseudo relevance feedback
- machine learning