Strategy Adaptation in Large Language Model Werewolf Agents
This work addresses the challenge of adaptive strategy selection in multi-agent conversational games like Werewolf, but it is incremental as it builds on prior prompt engineering methods.
The study tackled the problem of Werewolf agents' inability to adapt to changing game situations by proposing a method that explicitly switches between predefined strategies based on player attitudes and conversation context, resulting in improved performance compared to baseline agents with implicit or fixed strategies.
This study proposes a method to improve the performance of Werewolf agents by switching between predefined strategies based on the attitudes of other players and the context of conversations. While prior works of Werewolf agents using prompt engineering have employed methods where effective strategies are implicitly defined, they cannot adapt to changing situations. In this research, we propose a method that explicitly selects an appropriate strategy based on the game context and the estimated roles of other players. We compare the strategy adaptation Werewolf agents with baseline agents using implicit or fixed strategies and verify the effectiveness of our proposed method.