Enhancing Safety of Large Language Models via Embedding Space Separation

Documenti analoghi

• SEAS: Self-Evolving Adversarial Safety Optimization for Large Language Models
  di: Diao, Muxi, et al.
  Pubblicazione: (2024)
• Circumventing Safety Alignment in Large Language Models Through Embedding Space Toxicity Attenuation
  di: Zhang, Zhibo, et al.
  Pubblicazione: (2025)
• Multi-Perspective Consistency Enhances Confidence Estimation in Large Language Models
  di: Wang, Pei, et al.
  Pubblicazione: (2024)
• Compressing Sequences in the Latent Embedding Space: $K$-Token Merging for Large Language Models
  di: Xu, Zihao, et al.
  Pubblicazione: (2026)
• Safe-SAIL: Towards a Fine-grained Safety Landscape of Large Language Models via Sparse Autoencoder Interpretation Framework
  di: Weng, Jiaqi, et al.
  Pubblicazione: (2025)