Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment

Documenti analoghi

• PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization
  di: Cheng, Ruoxi, et al.
  Pubblicazione: (2024)
• SeCon-RAG: A Two-Stage Semantic Filtering and Conflict-Free Framework for Trustworthy RAG
  di: Si, Xiaonan, et al.
  Pubblicazione: (2025)
• Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Search
  di: Huang, Xun, et al.
  Pubblicazione: (2026)
• GeoShield: Safeguarding Geolocation Privacy from Vision-Language Models via Adversarial Perturbations
  di: Liu, Xinwei, et al.
  Pubblicazione: (2025)
• Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs
  di: Li, Jianan, et al.
  Pubblicazione: (2026)