Lockpicking LLMs: A Logit-Based Jailbreak Using Token-level Manipulation
Lockpicking LLMs: A Logit-Based Jailbreak Using Token-level Manipulation
About this item
Full title
Author / Creator
Li, Yuxi , Liu, Yi , Li, Yuekang , Shi, Ling , Deng, Gelei , Chen, Shengquan and Wang, Kailong
Publisher
Ithaca: Cornell University Library, arXiv.org
Journal title
Language
English
Formats
Publication information
Publisher
Ithaca: Cornell University Library, arXiv.org
Subjects
More information
Scope and Contents
Contents
Large language models (LLMs) have transformed the field of natural language processing, but they remain susceptible to jailbreaking attacks that exploit their capabilities to generate unintended and potentially harmful content. Existing token-level jailbreaking techniques, while effective, face scalability and efficiency challenges, especially as m...
Alternative Titles
Full title
Lockpicking LLMs: A Logit-Based Jailbreak Using Token-level Manipulation
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_3059626576
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_3059626576
Other Identifiers
E-ISSN
2331-8422
