Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Ruoyao, Todd, Graham, Xiao, Ziang, Yuan, Xingdi, Côté, Marc-Alexandre, Clark, Peter, Jansen, Peter
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2406.06485
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909220458201088
author Wang, Ruoyao
Todd, Graham
Xiao, Ziang
Yuan, Xingdi
Côté, Marc-Alexandre
Clark, Peter
Jansen, Peter
author_facet Wang, Ruoyao
Todd, Graham
Xiao, Ziang
Yuan, Xingdi
Côté, Marc-Alexandre
Clark, Peter
Jansen, Peter
contents Virtual environments play a key role in benchmarking advances in complex planning and decision-making tasks but are expensive and complicated to build by hand. Can current language models themselves serve as world simulators, correctly predicting how actions change different world states, thus bypassing the need for extensive manual coding? Our goal is to answer this question in the context of text-based simulators. Our approach is to build and use a new benchmark, called ByteSized32-State-Prediction, containing a dataset of text game state transitions and accompanying game tasks. We use this to directly quantify, for the first time, how well LLMs can serve as text-based world simulators. We test GPT-4 on this dataset and find that, despite its impressive performance, it is still an unreliable world simulator without further innovations. This work thus contributes both new insights into current LLM's capabilities and weaknesses, as well as a novel benchmark to track future progress as new models appear.
format Preprint
id arxiv_https___arxiv_org_abs_2406_06485
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Can Language Models Serve as Text-Based World Simulators?
Wang, Ruoyao
Todd, Graham
Xiao, Ziang
Yuan, Xingdi
Côté, Marc-Alexandre
Clark, Peter
Jansen, Peter
Computation and Language
Artificial Intelligence
Virtual environments play a key role in benchmarking advances in complex planning and decision-making tasks but are expensive and complicated to build by hand. Can current language models themselves serve as world simulators, correctly predicting how actions change different world states, thus bypassing the need for extensive manual coding? Our goal is to answer this question in the context of text-based simulators. Our approach is to build and use a new benchmark, called ByteSized32-State-Prediction, containing a dataset of text game state transitions and accompanying game tasks. We use this to directly quantify, for the first time, how well LLMs can serve as text-based world simulators. We test GPT-4 on this dataset and find that, despite its impressive performance, it is still an unreliable world simulator without further innovations. This work thus contributes both new insights into current LLM's capabilities and weaknesses, as well as a novel benchmark to track future progress as new models appear.
title Can Language Models Serve as Text-Based World Simulators?
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.06485