ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systems 事件

ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systems arXiv:2602.08567v2 Announce Type: replace-cross Abstract: Multi-agent large language model (LLM) systems increasingly consist of agents that observe and respond to one another's outputs. While value alignment is typically evaluated for isolated models, how value perturbations propagate through agent interactions remains poorly understood. We present ValueFlow, a perturbation-based framework that measures valu