From 367fc0eb679cf971570649f98e9dcff884fa6b1f Mon Sep 17 00:00:00 2001 From: Maxime MORGE <maxime.morge@univ-lille.fr> Date: Thu, 27 Mar 2025 21:05:06 +0100 Subject: [PATCH] LLM4AAMAS: Add a Game Theory Section --- README.md | 50 ++++++++++++++++++++++++++++---------------------- 1 file changed, 28 insertions(+), 22 deletions(-) diff --git a/README.md b/README.md index 6fe150f..68c2306 100644 --- a/README.md +++ b/README.md @@ -298,28 +298,6 @@ master, a designer or an analyst. Todd, Marvin Zammit, Sam Earle, Antonios Liapis, Julian Togelius, Georgios N. Yannakakis (2024) Published in *IEEE Transactions on Games* -LLMs have the ability to emulate a real human in certain experiments in -experimental economics or social psychology. - -- **[Large language models as simulated economic agents: What can we learn from - homo silicus?](https://www.nber.org/papers/w31122)** Horton, J. J. (2023). - National Bureau of Economic Research. - -LLMs, notably GPT-4 using ToT prompt, can simulate simple auction experiments in -line with theoretical expectations - -- *[The nuances of large-language-model-agent performance in simple English - auctions](https://www.academia.edu/download/112356998/13_231004_2023_Jan_Reg_Nuances_of_LLM_Performance_English_Auctions_Parady_USA_Published.pdf)** - Lamichhane, B., Palardy, J., & Singh, A. K. (2023). Empirical Economics - Letters,2(1). - -Generative consultants as economic agent with limited agency. - -- **[Generative AI as Economic - Agents](https://doi.org/10.1145/3699824.3699832)** Immorlica, N., Lucier, - B., & Slivkins, A. (2024). SIGecom Exch., 22(1), 93–109. ACM, New York, NY, - USA. - AGENTBENCH is a systematically designed multi-dimensional evolving benchmark for evaluating LLMs as agents which measures a significant performance gap between these top-tier models and their OSS competitors. @@ -361,6 +339,8 @@ simulation. Challenges](https://arxiv.org/abs/2402.01680)** Taicheng Guo et al. (2024) Published on *arXiv* arXiv:2402.01680 [cs.CL] +### Social Simulation + LLMs can simulate realistic perceptions, reasoning, and decision-making, react adaptively to environments without predefined explicit instructions by adjusting their responses through contextual learning mechanisms, autonomously generate @@ -416,6 +396,32 @@ A simulation of the propagation processes in a social network. Lu, Jinzhu Mao, Jinghua Piao, Huandong Wang, Depeng Jin, Yong Li (2023)* Published on *arXiv* arXiv:2307.14984 + +### Game Theory + + +LLMs have the ability to emulate a real human in certain experiments in +experimental economics or social psychology. + +- **[Large language models as simulated economic agents: What can we learn from + homo silicus?](https://www.nber.org/papers/w31122)** Horton, J. J. (2023). + National Bureau of Economic Research. + +LLMs, notably GPT-4 using ToT prompt, can simulate simple auction experiments in +line with theoretical expectations. + +- *[The nuances of large-language-model-agent performance in simple English + auctions](https://www.academia.edu/download/112356998/13_231004_2023_Jan_Reg_Nuances_of_LLM_Performance_English_Auctions_Parady_USA_Published.pdf)** + Lamichhane, B., Palardy, J., & Singh, A. K. (2023). Empirical Economics + Letters,2(1). + +Generative consultants as economic agent with limited agency. + +- **[Generative AI as Economic + Agents](https://doi.org/10.1145/3699824.3699832)** Immorlica, N., Lucier, + B., & Slivkins, A. (2024). SIGecom Exch., 22(1), 93–109. ACM, New York, NY, + USA. + This paper assesses the economic rationality of GPT's decisions across four domains: risk, time, social, and food preferences. The experiments reveal that GPT's decisions exhibit greater rationality than those of humans. This -- GitLab