Safe Multi-Agent Behavior Must Be Maintained, Not Merely Asserted: Constraint Drift in LLM-Based Multi-Agent Systems

Documenti analoghi

• Stop Reducing Responsibility in LLM-Powered Multi-Agent Systems to Local Alignment
  di: Hu, Jinwei, et al.
  Pubblicazione: (2025)
• Position: Towards a Responsible LLM-empowered Multi-Agent Systems
  di: Hu, Jinwei, et al.
  Pubblicazione: (2025)
• When KV Cache Reuse Fails in Multi-Agent Systems: Cross-Candidate Interaction is Crucial for LLM Judges
  di: Liang, Sichu, et al.
  Pubblicazione: (2026)
• Scalable Safe Multi-Agent Reinforcement Learning for Multi-Agent System
  di: Du, Haikuo, et al.
  Pubblicazione: (2025)
• How to Steer Your Multi-Agent System: Human-LLM Collaborative Planning
  di: He, Zeyu, et al.
  Pubblicazione: (2026)