arXiv cs.CY· Gaia Molinaro, Dave August, Danielle Perszyk, Anne G. E. Collins·· 12 小时前AI 评分44
研究发现 LLM 在自主导向学习任务中的目标选择与人类存在显著差异
Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task
AI 导读
一项认知科学自主导向学习任务测试显示，GPT-5、Gemini 2.5 Pro、Claude Sonnet 4.5、Qwen3 32B 和 Centaur 五个模型的目标选择行为与人类存在显著差异：人类会逐步探索并形成个体多样化的目标路径，多数模型则只利用单一已识别解法，或表现异常低下，且同一模型的多次实例间差异很小。
正文 · 原文
Abstract: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.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY) |
| Cite as: | arXiv:2603.03295 [cs.CL] |
| (or arXiv:2603.03295v3 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2603.03295 arXiv-issued DOI via DataCite |
Submission history
From: Gaia Molinaro [view email]
[v1]
Fri, 6 Feb 2026 15:39:54 UTC (4,022 KB)
[v2]
Wed, 13 May 2026 10:48:07 UTC (14,886 KB)
[v3]
Fri, 16 Oct 2026 03:16:35 UTC (21,262 KB)
来源：arXiv cs.CY · arxiv.org