UOR: Universal Backdoor Attacks on Pre-trained Language Models.
Wei DuPeixuan LiBoqun LiHaodong ZhaoGongshen LiuPublished in: CoRR (2023)
Keyphrases
- language model
- pre trained
- language modeling
- training data
- training examples
- speech recognition
- probabilistic model
- n gram
- language modelling
- information retrieval
- document retrieval
- retrieval model
- test collection
- query expansion
- statistical language models
- appearance variations
- control signals
- machine learning
- information extraction
- active learning
- relevance model
- document ranking
- multimedia
- smoothing methods
- feature extraction
- language models for information retrieval