Attributing and Exploiting Safety Vectors through Global Optimization in Large Language Models

Documenti analoghi

• Knowledge-Driven Multi-Turn Jailbreaking on Large Language Models
  di: Li, Songze, et al.
  Pubblicazione: (2026)
• TooBadRL: Trigger Optimization to Boost Effectiveness of Backdoor Attacks on Deep Reinforcement Learning
  di: Zhang, Mingxuan, et al.
  Pubblicazione: (2025)
• ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models
  di: Xie, Jin, et al.
  Pubblicazione: (2025)
• Unveiling the Security Risks of Federated Learning in the Wild: From Research to Practice
  di: Chen, Jiahao, et al.
  Pubblicazione: (2026)
• Towards Scalable and Interpretable Mobile App Risk Analysis via Large Language Models
  di: Yang, Yu, et al.
  Pubblicazione: (2025)