Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models.
Jiashu XuMingyu Derek MaFei WangChaowei XiaoMuhao ChenPublished in: CoRR (2023)
Keyphrases
- language model
- sat instances
- language modeling
- n gram
- speech recognition
- test collection
- information retrieval
- retrieval model
- document retrieval
- probabilistic model
- statistical language models
- query expansion
- query terms
- context sensitive
- sat solvers
- satisfiability problem
- language models for information retrieval
- vector space model
- multimedia
- pseudo relevance feedback
- ad hoc information retrieval
- sat problem
- smoothing methods
- language modelling
- machine learning
- retrieval effectiveness
- information extraction
- instruction set
- feature selection
- okapi bm