1 paper · 1 filter
Yihuai Lan, Zhiqiang Hu, Lei Wang +6
This paper explores the open research problem of understanding the social behaviors of LLM-based agents. Using Avalon as a testbed, we employ system prompts to guide LLM agents in…