Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2310.14985 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910646981885952 |
|---|---|
| author | Lan, Yihuai Hu, Zhiqiang Wang, Lei Wang, Yang Ye, Deheng Zhao, Peilin Lim, Ee-Peng Xiong, Hui Wang, Hao |
| author_facet | Lan, Yihuai Hu, Zhiqiang Wang, Lei Wang, Yang Ye, Deheng Zhao, Peilin Lim, Ee-Peng Xiong, Hui Wang, Hao |
| contents | 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 gameplay. While previous studies have touched on gameplay with LLM agents, research on their social behaviors is lacking. We propose a novel framework, tailored for Avalon, features a multi-agent system facilitating efficient communication and interaction. We evaluate its performance based on game success and analyze LLM agents' social behaviors. Results affirm the framework's effectiveness in creating adaptive agents and suggest LLM-based agents' potential in navigating dynamic social interactions. By examining collaboration and confrontation behaviors, we offer insights into this field's research and applications. Our code is publicly available at https://github.com/3DAgentWorld/LLM-Game-Agent. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_14985 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon Gameplay Lan, Yihuai Hu, Zhiqiang Wang, Lei Wang, Yang Ye, Deheng Zhao, Peilin Lim, Ee-Peng Xiong, Hui Wang, Hao Computation and Language 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 gameplay. While previous studies have touched on gameplay with LLM agents, research on their social behaviors is lacking. We propose a novel framework, tailored for Avalon, features a multi-agent system facilitating efficient communication and interaction. We evaluate its performance based on game success and analyze LLM agents' social behaviors. Results affirm the framework's effectiveness in creating adaptive agents and suggest LLM-based agents' potential in navigating dynamic social interactions. By examining collaboration and confrontation behaviors, we offer insights into this field's research and applications. Our code is publicly available at https://github.com/3DAgentWorld/LLM-Game-Agent. |
| title | LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon Gameplay |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2310.14985 |