{"id":"doi:10.1109/tcds.2024.3377445","kind":"article","title":"The Inadequacy of Reinforcement Learning From Human Feedback—Radicalizing Large Language Models via Semantic Vulnerabilities","authors":["Timothy R. McIntosh","Teo Sušnjak","Tong Liu","Paul Watters","Malka N. Halgamuge"],"author_count":5,"year":2024,"published":"2024-03-18","venue":"IEEE Transactions on Cognitive and Developmental Systems","cited_by":61,"doi":"10.1109/tcds.2024.3377445","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1109/tcds.2024.3377445","archive_url":"https://space-kitty.com/papers/doi:10.1109/tcds.2024.3377445","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W4392908117","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"retracted":false,"peer_reviewed":null,"categories":["Ethics and Social Impacts of AI"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0).","topics":[{"slug":"alignment-safety","name":"AI Alignment & Safety","domain_slug":"ai"}],"cited_in_dossiers":[]}