The Inadequacy of Reinforcement Learning From Human Feedback—Radicalizing Large Language Models via Semantic Vulnerabilities
Timothy R. McIntosh, Teo Sušnjak, Tong Liu, Paul Watters, Malka N. Halgamuge
The abstract is not reproduced here because its license does not permit redistribution. Read it at the original source.
Record
- Archive ID
- doi:10.1109/tcds.2024.3377445
- Published
- Mar 18, 2024
- Venue
- IEEE Transactions on Cognitive and Developmental Systems
- DOI
- 10.1109/tcds.2024.3377445
- Categories
- Ethics and Social Impacts of AI
- Provenance
- openalex