GUARD: Role-playing to Generate Natural-language Jailbreakings to Test Guideline Adherence of Large Language Models.
Haibo JinRuoxi ChenAndy ZhouJinyin ChenYang ZhangHaohan WangPublished in: CoRR (2024)
Keyphrases
- language model
- role playing
- natural language
- language modeling
- n gram
- probabilistic model
- document retrieval
- information retrieval
- statistical language models
- speech recognition
- retrieval model
- test collection
- context sensitive
- ad hoc information retrieval
- language modelling
- machine learning
- vector space model
- relevance model
- query expansion
- smoothing methods
- translation model
- query terms
- language processing
- cross lingual
- natural language processing
- information extraction
- question answering
- language models for information retrieval
- language model for information retrieval
- passage retrieval
- document ranking
- word error rate
- hidden markov models