A Win-win Deal: Towards Sparse and Robust Pre-trained Language Models.
Yuanxin LiuFandong MengZheng LinJiangnan LiPeng FuYanan CaoWeiping WangJie ZhouPublished in: CoRR (2022)
Keyphrases
- language model
- pre trained
- language modeling
- n gram
- probabilistic model
- win win
- information retrieval
- query expansion
- retrieval model
- speech recognition
- language models for information retrieval
- smoothing methods
- statistical language models
- relevance model
- training data
- neural network
- dynamic environments
- small number
- active learning
- partial occlusion
- illumination variations
- face recognition
- data mining
- data sets