# 智能体探索却忽视：LLM缺乏环境好奇心

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-04-19 08:00
- AIHOT 链接：https://aihot.virxact.com/items/cmo8o3iev069pslml4lt8tlcd
- 原文链接：https://arxiv.org/abs/2604.17609

## AI 摘要

研究发现当前LLM智能体缺乏"环境好奇心"，即识别并利用环境中意外相关信息的能力。研究者在Terminal-Bench等三个基准测试中注入完整解决方案：Terminal-Bench中智能体79-81%发现方案但仅37-50%利用；AppWorld中超90%看到标注"返回完整解决方案"的文档却不足7%利用。工具配置、测试时计算和训练数据分布是三大影响因素。即便优化配置，智能体仍在多数试验中忽视已发现方案，仅将环境用于获取预期信息而非调整策略。

## 正文

LLM-based agents are assumed to integrate environmental observations into their reasoning: discovering highly relevant but unexpected information should naturally lead to a model exploiting its own discoveries. We show that this assumption is false for current LLM-based agents, which struggle to reflect or react to unexpected information. Across three benchmarks (Terminal-Bench, SWE-Bench, AppWorld), we inject complete task solutions into the agent environments to deliberately expose a task's solution to a model. While agents discover these solutions on Terminal-Bench in 79-81% of runs, they interact, or exploit, them in only 37-50% of cases. This gap is starkest in AppWorld: agents see documentation stating that a command "returns the complete solution to this task" in over 90% of attempts but exploit this in fewer than 7% of trials. We show that agents lack what we call environmental curiosity: the capability to recognize and investigate unexpected but relevant observations in response to environmental stimuli. We identify three main factors influencing environmental curiosity: available tools in the agent scaffold, test-time compute, and training data distribution. Our findings identify configurations that maximize curiosity also achieve the best performance on the unmodified benchmarks. Yet even jointly optimized agents still ignore discovered solutions in the majority of trials: current agents use the environment to fetch expected information, but not to revise their strategy or maximally exploit useful stimuli.