论文：LLM 智能体的采样不是好奇心，应建立可在线更新的环境预测模型
Sampling Is Not Curiosity: Why LLM Agents Should Investigate
搞 Agent 的都该看看这篇。它说清了为什么加了 RAG、树搜索的智能体还是学不到东西——因为权重不动。想做出能真正从交互里学习的探索型 Agent，这篇给了定义和路线图。
arXiv 论文 2611.10501 提出，当前 LLM 智能体在开放任务中做的是随机探索而非好奇心驱动探索。作者指出，推理阶段的 LLM 是从上下文到下一 token 分布的固定映射，temperature、best-of-N、树搜索、RAG 等机制只改变采样方式或条件上下文，不会更新智能体对环境的模型。
论文借鉴 RL 文献中对好奇心的定义，给出四个判据：拥有对环境响应的预测结构、能从当前交互在线改进、通过预期改进驱动动作选择、并在上下文窗口之外持续存在。开放场景下还要求智能体能自主决定调查什么。据此作者提出一个框架，把"好奇心"一词保留给满足这些条件的系统，并列出实现该框架可展开的研究方向。
这与把探索能力等同于提高采样多样性的常见做法形成对比，对构建科研智能体和长程自主智能体的开发者有直接参考意义。
Sampling Is Not Curiosity: Why LLM Agents Should Investigate
LLM agents are increasingly used for open-ended tasks, very common in research scenarios where they are required to explore, investigate, or act curiously. At inference time, an LLM is a fixed map from context to next-token distribution, the pretrained weights determine the function, and only its conditioning changes as context accumulates. Mechanisms like temperature, best-of-N, tree search, or RAG vary how the distribution is sampled or what it is conditioned on, but do not update the agent's model of the environment. We argue that current LLM agents do not explore curiously; they explore stochastically. From how curiosity is understood in the RL literature, it requires a predictive structure over how the environment responds to the agent's actions, which improves online from current interaction, drives action selection through its predicted improvement, and persists beyond the context window. Additionally, in its open-ended form, the agent should intrinsically originate what to investigate. Thus, we propose a framework defining curiosity in LLM agents, reserving the term for systems that satisfy it. In doing so, we identify the research directions its implementation opens up.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include LLM智能体, 好奇心, 强化学习, arXiv.
Source attribution: curated from arXiv cs.AI (@Alfonso Amayuelas, Piotr Piękos, Xin Eric Wang, William Yang Wang, Jürgen Schmidhuber, Alex Pentland).
