1 paper · 1 filter
Jinnuo Liu, Chuke Liu, Hua Shen
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 is…