diff --git a/doc/paper/ICTAI25/abstract.txt b/doc/paper/ICTAI25/abstract.txt index c640a008026a87f6723756d9bd8bb6e656eb8d3b..eb7579fe4e048e84be3cea19a113f8aea5fddebd 100644 --- a/doc/paper/ICTAI25/abstract.txt +++ b/doc/paper/ICTAI25/abstract.txt @@ -1,9 +1 @@ -Recent advances in Large Language Models (LLMs) have enabled the creation of -Generative Agents (GAs) capable of autonomous decision-making in interactive -settings. This paper investigates whether GAs can exhibit socially credible -behavior. Drawing from behavioral game theory, we evaluate five state-of-the-art -models across three canonical game-theoretic environments. Our results show that -while some GAs can accurately predict their opponent’s behavior, few are able to -incorporate those predictions into decision-making. These behavioral flaws help -explain why coordination remains especially challenging: most models struggle to -align with others, even when communication is allowed. +Recent advances in Large Language Models (LLMs) have enabled the creation of Generative Agents (GAs) capable of autonomous decision-making in interaction. This paper investigates whether GAs can exhibit socially credible behavior. Drawing from behavioral game theory, we evaluate five state-of-the-art models across three canonical game-theoretic environments. Our results show that, while some GAs can accurately predict their opponent’s behavior, few are able to incorporate those predictions into decision-making. These behavioral flaws help explain why coordination remains especially challenging: most models struggle to align with others, even when communication is allowed.