SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models.
Wei HuangHaotong QinYangdong LiuYawei LiXianglong LiuLuca BeniniMichele MagnoXiaojuan QiPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- probabilistic model
- n gram
- speech recognition
- document retrieval
- information retrieval
- language modelling
- query expansion
- vector space model
- test collection
- language models for information retrieval
- context sensitive
- statistical language models
- average precision
- smoothing methods
- precision and recall
- ad hoc information retrieval
- retrieval model
- query terms
- pseudo relevance feedback
- image annotation
- document ranking
- term dependencies
- word error rate
- language model for information retrieval