Agentic crafting requires LLMs to operate in real-world environments over multiple turns by taking actions, observing outcomes, and iteratively refining artifacts. Despite its importance, the open-source community lacks a principled, end-to-end ecosystem to streamline agent development. We introduce the Agentic Learning Ecosystem (ALE), a foundational infrastructure that optimizes the production pipeline for agentic model. ALE consists of three components: ROLL, a post-training framework for weight optimization; ROCK, a sandbox environment manager for trajectory generation; and iFlow CLI, an agent framework for efficient context engineering. We release ROME, an open-source agent grounded by ALE and trained on over one million trajectories. Our approach includes data composition protocols for synthesizing complex behaviors and a novel policy optimization algorithm, Interaction-Perceptive Agentic Policy Optimization (IPA), which assigns credit over semantic interaction chunks rather than individual tokens to improve long-horizon training stability. Empirically, we evaluate ROME within a structured setting and introduce Terminal Bench Pro, a benchmark with improved scale and contamination control. ROME demonstrates strong performance across benchmarks like SWE-bench Verified and Terminal Bench, proving the effectiveness of ALE.

翻译：智能体构建要求大型语言模型在现实环境中通过多轮操作执行动作、观察结果并迭代优化产出物。尽管其重要性日益凸显，开源社区仍缺乏一个系统化、端到端的生态系统来简化智能体开发流程。本文提出智能体学习生态系统（ALE），这是一个优化智能体模型生产流程的基础设施。ALE包含三个核心组件：ROLL（用于权重优化的后训练框架）、ROCK（用于轨迹生成的沙盒环境管理器）以及iFlow CLI（用于高效上下文工程的智能体框架）。我们发布了基于ALE构建并经过超百万条轨迹训练的开源智能体ROME。该方法包含用于合成复杂行为的数据组合协议，以及新颖的策略优化算法——交互感知智能体策略优化（IPA）。该算法通过在语义交互片段而非单个令牌层面进行信用分配，显著提升了长周期训练的稳定性。我们在结构化环境中对ROME进行实证评估，并推出具有更优规模与污染控制特性的基准测试集Terminal Bench Pro。实验表明，ROME在SWE-bench Verified和Terminal Bench等基准测试中均表现出色，验证了ALE系统的有效性。