GUARD-SLM: Token Activation-Based Defense Against Jailbreak Attacks for Small Language Models Preprint

Mia, Md Jueal, Molto, Joaquin, Wu, Yanzhao et al. (2026). GUARD-SLM: Token Activation-Based Defense Against Jailbreak Attacks for Small Language Models . 10.48550/arxiv.2603.28817

cited authors

  • Mia, Md Jueal; Molto, Joaquin; Wu, Yanzhao; Amini, M Hadi

authors

publication date

  • March 28, 2026

keywords

  • 46 Information and Computing Sciences
  • 4604 Cybersecurity and Privacy

Digital Object Identifier (DOI)