ProxyWar: Dynamic Assessment of LLM Code Generation in Game Arenas

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Peng, Wenjun, Wang, Xinyu, Wu, Qi
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917247142854656
author Peng, Wenjun
Wang, Xinyu
Wu, Qi
author_facet Peng, Wenjun
Wang, Xinyu
Wu, Qi
contents Large language models (LLMs) have revolutionized automated code generation, yet the evaluation of their real-world effectiveness remains limited by static benchmarks and simplistic metrics. We present ProxyWar, a novel framework that systematically assesses code generation quality by embedding LLM-generated agents within diverse, competitive game environments. Unlike existing approaches, ProxyWar evaluates not only functional correctness but also the operational characteristics of generated programs, combining automated testing, iterative code repair, and multi-agent tournaments to provide a holistic view of program behavior. Applied to a range of state-of-the-art coders and games, our approach uncovers notable discrepancies between benchmark scores and actual performance in dynamic settings, revealing overlooked limitations and opportunities for improvement. These findings highlight the need for richer, competition-based evaluation of code generation. Looking forward, ProxyWar lays a foundation for research into LLM-driven algorithm discovery, adaptive problem solving, and the study of practical efficiency and robustness, including the potential for models to outperform hand-crafted agents. The project is available at https://github.com/xinke-wang/ProxyWar.
format Preprint
id arxiv_https___arxiv_org_abs_2602_04296
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ProxyWar: Dynamic Assessment of LLM Code Generation in Game Arenas
Peng, Wenjun
Wang, Xinyu
Wu, Qi
Software Engineering
Artificial Intelligence
Large language models (LLMs) have revolutionized automated code generation, yet the evaluation of their real-world effectiveness remains limited by static benchmarks and simplistic metrics. We present ProxyWar, a novel framework that systematically assesses code generation quality by embedding LLM-generated agents within diverse, competitive game environments. Unlike existing approaches, ProxyWar evaluates not only functional correctness but also the operational characteristics of generated programs, combining automated testing, iterative code repair, and multi-agent tournaments to provide a holistic view of program behavior. Applied to a range of state-of-the-art coders and games, our approach uncovers notable discrepancies between benchmark scores and actual performance in dynamic settings, revealing overlooked limitations and opportunities for improvement. These findings highlight the need for richer, competition-based evaluation of code generation. Looking forward, ProxyWar lays a foundation for research into LLM-driven algorithm discovery, adaptive problem solving, and the study of practical efficiency and robustness, including the potential for models to outperform hand-crafted agents. The project is available at https://github.com/xinke-wang/ProxyWar.
title ProxyWar: Dynamic Assessment of LLM Code Generation in Game Arenas
topic Software Engineering
Artificial Intelligence
url https://arxiv.org/abs/2602.04296