Summary of Helmsman Of the Masses? Evaluate the Opinion Leadership Of Large Language Models in the Werewolf Game, by Silin Du et al.
Helmsman of the Masses? Evaluate the Opinion Leadership of Large Language Models in the Werewolf Game
by Silin Du, Xiaowei Zhang
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper investigates the opinion leadership abilities of Large Language Models (LLMs) in social deductive games. The researchers employ the Werewolf game as a simulation platform to assess the opinion leadership of LLM-based agents, focusing on the Sheriff role that summarizes arguments and recommends decision options. To evaluate the opinion leaders, the authors develop two novel metrics: reliability and influence. They conduct extensive experiments with LLMs of different scales, collect a Werewolf question-answering dataset (WWQA) to enhance LLM’s grasp of game rules, and incorporate human participants for further analysis. The results show that few LLMs possess the capacity for opinion leadership. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study looks at how well artificial intelligence (AI) can lead opinions in a group setting. They used a game called Werewolf to test their AI models. In this game, there are different roles like the Sheriff who has to summarize arguments and make recommendations. The researchers created two new ways to measure how good an opinion leader is: reliability and influence. They tested their AI models on the game and collected data from human players too. The results show that most AI models aren’t very good at leading opinions. |
Keywords
» Artificial intelligence » Question answering
