大型语言模型思考得太快，无法有效地进行探索。

Large Language Models Think Too Fast To Explore Effectively

January 29, 2025

作者: Lan Pan, Hanbo Xie, Robert C. Wilson

cs.AI

摘要

大型语言模型已经展现出许多智能能力。尽管有许多基准评估它们的智能，但对它们的探索能力却付之闻言，而探索能力对于在自然和人工系统中发现新信息并适应新环境至关重要。LLM能够在开放式任务中有效探索的程度，特别是在开放式任务中，仍然不清楚。本研究调查了LLM在开放式任务中是否能够超越人类的探索能力，以Little Alchemy 2作为范例，代理通过组合元素来发现新元素。结果显示，除了o1模型外，大多数LLM在探索方面表现不及人类，这些传统LLM主要依赖于不确定性驱动的策略，而不像人类那样平衡不确定性和赋权。通过对稀疏自动编码器的模型进行表征分析，发现不确定性和选择在较早的变压器块中得到了表征，而赋权值则在后期处理，导致LLM思考过快并做出过早决策，阻碍了有效的探索。这些发现揭示了LLM探索的局限性，并提出了改善它们适应性的方向。

English

Large Language Models have emerged many intellectual capacities. While numerous benchmarks assess their intelligence, limited attention has been given to their ability to explore, an essential capacity for discovering new information and adapting to novel environments in both natural and artificial systems. The extent to which LLMs can effectively explore, particularly in open-ended tasks, remains unclear. This study investigates whether LLMs can surpass humans in exploration during an open-ended task, using Little Alchemy 2 as a paradigm, where agents combine elements to discover new ones. Results show most LLMs underperform compared to humans, except for the o1 model, with those traditional LLMs relying primarily on uncertainty driven strategies, unlike humans who balance uncertainty and empowerment. Representational analysis of the models with Sparse Autoencoders revealed that uncertainty and choices are represented at earlier transformer blocks, while empowerment values are processed later, causing LLMs to think too fast and make premature decisions, hindering effective exploration. These findings shed light on the limitations of LLM exploration and suggest directions for improving their adaptability.