Results 161 to 170 of about 11,592 (238)
Some of the next articles are maybe not open access.
Language Agents with Reinforcement Learning for Strategic Play in the Werewolf Game
International Conference on Machine Learning, 2023Agents built with large language models (LLMs) have shown great potential across a wide range of domains. However, in complex decision-making tasks, pure LLM-based agents tend to exhibit intrinsic bias in their choice of actions, which is inherited from ...
Zelai Xu +4 more
semanticscholar +1 more source
2020
The judicial bestiary at the heart of eighteenth-century politics has long been evident in Enlightenment social contract debates, as Michel Foucault’s and Giorgio Agamben’s theories of biopolitics show. In this essay, I argue that Wollstonecraft is nonetheless the first thinker of ‘true’ werewolf out-lawry in her final novel, Maria, Or the Wrongs of ...
openaire +1 more source
The judicial bestiary at the heart of eighteenth-century politics has long been evident in Enlightenment social contract debates, as Michel Foucault’s and Giorgio Agamben’s theories of biopolitics show. In this essay, I argue that Wollstonecraft is nonetheless the first thinker of ‘true’ werewolf out-lawry in her final novel, Maria, Or the Wrongs of ...
openaire +1 more source
Enhance Reasoning for Large Language Models in the Game Werewolf
arXiv.orgThis paper presents an innovative framework that integrates Large Language Models (LLMs) with an external Thinker module to enhance the reasoning capabilities of LLM-based agents. Unlike augmenting LLMs with prompt engineering, Thinker directly harnesses
Shuang Wu +6 more
semanticscholar +1 more source
Werewolf Arena: A Case Study in LLM Evaluation via Social Deduction
arXiv.orgThis paper introduces Werewolf Arena, a novel framework for evaluating large language models (LLMs) through the lens of the classic social deduction game, Werewolf.
Suma Bailis +2 more
semanticscholar +1 more source
International Conference on Machine Learning
Large language model (LLM) agents have recently demonstrated impressive capabilities in various domains like open-ended conversation and multi-step decision-making.
Zelai Xu +4 more
semanticscholar +1 more source
Large language model (LLM) agents have recently demonstrated impressive capabilities in various domains like open-ended conversation and multi-step decision-making.
Zelai Xu +4 more
semanticscholar +1 more source
International Conference on Foundations of Digital Games
In this study, we explore the reasoning capabilities of Large Language Models (LLMs) within the context of the social communication game Werewolf, aiming to evaluate their performance in managing complex system states commonly found in computer games ...
Christian Poglitsch +2 more
semanticscholar +1 more source
In this study, we explore the reasoning capabilities of Large Language Models (LLMs) within the context of the social communication game Werewolf, aiming to evaluate their performance in managing complex system states commonly found in computer games ...
Christian Poglitsch +2 more
semanticscholar +1 more source
WOLF: Werewolf-based Observations for LLM Deception and Falsehoods
arXiv.orgDeception is a fundamental challenge for multi-agent reasoning: effective systems must strategically conceal information while detecting misleading behavior in others.
Mrinal Agarwal +7 more
semanticscholar +1 more source
MultiMind: Enhancing Werewolf Agents with Multimodal Reasoning and Theory of Mind
ACM MultimediaLarge Language Model (LLM) agents have demonstrated impressive capabilities in social deduction games (SDGs) like Werewolf, where strategic reasoning and social deception are essential.
Zheng Zhang +4 more
semanticscholar +1 more source
Learning to Discuss Strategically: A Case Study on One Night Ultimate Werewolf
Neural Information Processing SystemsCommunication is a fundamental aspect of human society, facilitating the exchange of information and beliefs among people. Despite the advancements in large language models (LLMs), recent agents built with these often neglect the control over discussion ...
Xuanfa Jin +5 more
semanticscholar +1 more source