LLMSafeGuard: A Training-Free Framework for Safeguarding LLM Decoding via Context-Wise Similarity Validation
Open Access
- 5 July 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1713-1720
Abstract
No abstract availableThis publication has 10 references indexed in Scilit:
- GradSafe: Detecting Jailbreak Prompts for LLMs via Safety-Critical Gradient Analysis2024
- How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs2024
- Defending ChatGPT against jailbreak attack via self-remindersNature Machine Intelligence, 2023
- A Survey of Controllable Text Generation Using Transformer-based Pre-trained Language ModelsACM Computing Surveys, 2023
- Recent Advances in Natural Language Processing via Large Pre-trained Language Models: A SurveyACM Computing Surveys, 2023
- Copyright Violations and Large Language Models2023
- Controllable Natural Language Generation with Contrastive Prefixes2022
- Measuring and Improving Consistency in Pretrained Language ModelsTransactions of the Association for Computational Linguistics, 2021
- DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts2021
- FUDGE: Controlled Text Generation With Future Discriminators2021