QuickLLaMA: Query-aware Inference Acceleration for Large Language Models.
Jingyao LiHan ShiXin JiangZhenguo LiHong XuJiaya JiaPublished in: CoRR (2024)
Keyphrases
- language model
- query expansion
- query terms
- query specific
- document ranking
- query generation
- language modeling
- relevance model
- term dependencies
- n gram
- statistical language modeling
- probabilistic model
- language modelling
- original query
- pseudo feedback
- document retrieval
- retrieval model
- document length
- ir models
- information retrieval
- query processing
- retrieval effectiveness
- pseudo relevance feedback
- statistical language models
- speech recognition
- spoken term detection
- probabilistic retrieval models
- vector space model
- context sensitive
- language models for information retrieval
- language modeling framework
- test collection
- bayesian networks
- sentence retrieval
- user queries
- relevant documents
- translation model
- document collections
- information retrieval systems
- smoothing methods
- relevance feedback
- passage retrieval
- improve retrieval effectiveness
- inter document similarities
- average precision