FlashDecoding++: Faster Large Language Model Inference with Asynchronization, Flat GEMM Optimization, and Heuristics.
Ke HongGuohao DaiJiaming XuQiuli MaoXiuhong LiJun LiuKangdi ChenYuhan DongYu WangPublished in: MLSys (2024)
Keyphrases
- language model
- language modeling
- document retrieval
- n gram
- speech recognition
- probabilistic model
- information retrieval
- language modelling
- statistical language models
- context sensitive
- query expansion
- retrieval model
- smoothing methods
- language model for information retrieval
- ad hoc information retrieval
- test collection
- mixture model
- vector space model
- document ranking
- pseudo relevance feedback
- query terms
- bayesian networks
- cross lingual
- dependency structure
- document length
- word error rate
- generative model