LLM Multi-Agent Systems Are Vulnerable to Herding and Runaway Cascades
Resumo indisponível. Consulte o paper original.
PAPER / ARXIV:2610.14800
Yuchen Ma, Swadheen Shukla, Paulo Salem, Yue Huang, Robert Sim, Stefan Feuerriegel, Michel Galley, Jianfeng Gao
RESUMO
LLM agents increasingly divide enterprise work across roles with different goals, authority, access, and information. Because each agent often sees only a local snapshot of the enterprise workflow, locally reasonable actions can still violate organization-wide constraints (e.g., several service agents can each issue credit that is individually permitted while collectively exceeding a shared compensation cap). We introduce EAIR, the first framework for analyzing enterprise multi-agent interaction risks: we formalize enterprise interaction risk as a taxonomy of testable risks, build EAIR-Worlds, a suite of executable micro-worlds instantiating selected risks as controlled workflows, and develop a diagnosis-and-repair pipeline that locates contributing interaction boundaries and guides targeted workflow repairs. Large-scale experiments show recurring interaction failures across the organizational properties studied.
NO MESMO MAPA
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.
Resumo indisponível. Consulte o paper original.