AI 中文总结
该研究针对多模态智能体学习中单纯扩展环境数量无效的问题，提出AES与HDC两种方法优化环境分布，提升了智能体训练效果。
AI 中文摘要
近期研究通过构建大规模多模态环境池来训练智能体，但我们发现单纯增加多模态环境的数量并不总能带来增益。我们通过一系列实验分析了当前多模态环境分布的局限性，基于这些发现，从**多样性**和**难度结构**两个维度研究如何构建更有效的训练环境分布。在多样性方面，我们提出**能力感知环境选择（AES）**以获取多样的环境集；在难度结构方面，我们提出**分层难度课程（HDC）**，通过弱化利用和状态规模递进两个难度级别组织课程学习。实验表明，AES和HDC可有效提升多模态智能体的训练效果。
英文摘要
Recent works train agents by constructing large-scale multimodal environment pools. However, we find that simply increasing the number of multimodal environments does not always benefit. We further analyze the limitations in current multimodal environment distributions through a series of experiments. Based on these findings, we study how to build more effective training environment distributions from two dimensions: **diversity** and **difficulty structure**. For diversity, we propose **Ability-aware Environment Selection (AES)** to obtain diverse environment sets. For difficulty structure, we propose **Hierarchical Difficulty Curriculum (HDC)**, which organizes curriculum learning through two difficulty levels: harness weakening and state-scale progression. Experiments show that AES and HDC effectively improve multimodal agent training.
CommentsCode: https://github.com/GaryStack/Beyond-MMEnv-Scaling
