The Inadequacy of Reinforcement Learning From Human Feedback - Radicalizing Large Language Models via Semantic Vulnerabilities.
Timothy R. McIntoshTeo SusnjakTong LiuPaul A. WattersMalka N. HalgamugePublished in: IEEE Trans. Cogn. Dev. Syst. (2024)
Keyphrases
- language model
- reinforcement learning
- language modeling
- probabilistic model
- n gram
- language modelling
- document retrieval
- information retrieval
- speech recognition
- query expansion
- retrieval model
- test collection
- smoothing methods
- context sensitive
- language models for information retrieval
- ad hoc information retrieval
- document ranking
- language model for information retrieval
- statistical language models
- query terms
- vector space model
- semantic information
- natural language
- user feedback
- retrieval effectiveness
- text retrieval
- relevance feedback