Guardado en:
Detalles Bibliográficos
Autores principales: Molinaro, Gaia, August, Dave, Perszyk, Danielle, Collins, Anne G. E.
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2603.03295
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911678633869312
author Molinaro, Gaia
August, Dave
Perszyk, Danielle
Collins, Anne G. E.
author_facet Molinaro, Gaia
August, Dave
Perszyk, Danielle
Collins, Anne G. E.
contents Whether in agentic workflows, social studies, or chat settings, large language models (LLMs) are increasingly being asked to replace humans in choosing which goals to pursue, rather than completing predefined tasks. However, the assumption that LLMs accurately reflect human preferences for goal setting remains largely untested. We assess the validity of LLMs as proxies for human goal selection in a controlled, self-directed learning task borrowed from cognitive science. Across five models (GPT-5, Gemini 2.5 Pro, Claude Sonnet 4.5, Qwen3 32B, and Centaur), we find substantial divergence from human behavior. While people gradually explore and learn to achieve goals with diversity across individuals, most models exploit a single identified solution or show surprisingly low performance, with distinct patterns across models and little variability across instances of the same model. Chain-of-thought reasoning and persona steering provide limited improvements, and our conclusions hold across experimental settings. While they await confirmation in applied settings, these findings highlight the uniqueness of human goal selection and caution against its replacement with current models.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03295
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task
Molinaro, Gaia
August, Dave
Perszyk, Danielle
Collins, Anne G. E.
Computation and Language
Artificial Intelligence
Computers and Society
Whether in agentic workflows, social studies, or chat settings, large language models (LLMs) are increasingly being asked to replace humans in choosing which goals to pursue, rather than completing predefined tasks. However, the assumption that LLMs accurately reflect human preferences for goal setting remains largely untested. We assess the validity of LLMs as proxies for human goal selection in a controlled, self-directed learning task borrowed from cognitive science. Across five models (GPT-5, Gemini 2.5 Pro, Claude Sonnet 4.5, Qwen3 32B, and Centaur), we find substantial divergence from human behavior. While people gradually explore and learn to achieve goals with diversity across individuals, most models exploit a single identified solution or show surprisingly low performance, with distinct patterns across models and little variability across instances of the same model. Chain-of-thought reasoning and persona steering provide limited improvements, and our conclusions hold across experimental settings. While they await confirmation in applied settings, these findings highlight the uniqueness of human goal selection and caution against its replacement with current models.
title Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task
topic Computation and Language
Artificial Intelligence
Computers and Society
url https://arxiv.org/abs/2603.03295