这篇论文用OpenAI Codex的思路做智能体技能提取，八个聚类五个纯度超0.95，但GRPO只提了2个点，分析很实在。
论文提出三阶段流水线，从GUI轨迹中分段、聚类候选技能并训练技能感知策略。八个聚类中五个纯度≥0.95。但GRPO仅将技能步准确率从18.5%提升至20.5%，低于频率先验。作者指出弱边界检测器、无序段表示和离线奖励模型是三大原因。
// Automating SKILL.md Generation // Increasingly, mining sessions is one of the best ways to impro...
// Automating SKILL.md Generation // Increasingly, mining sessions is one of the best ways to improve your agents. OpenAI released something similar yesterday that lets Codex package skills from interactions. (bookmark it) This paper explains a related approach. They run a three-stage pipeline that segments GUI trajectories, clusters them into candidate skills, and trains a skill-aware policy. The clusters are genuinely readable, with five of eight hitting 0.95 or higher purity against ground-truth workflow labels. But readability does not transfer. GRPO lifts skill-step accuracy only from 18.5% to 20.5%, leaves BrowseComp+ flat, and loses to trivial frequency priors. The authors name the three culprits: a weak boundary detector, an orderless segment representation, and an offline reward model. Paper: arxiv.org/abs/2606.20363 Learn to build effective AI agents in our academy: academy.dair.ai 💬 4 🔄 2 ❤️ 12 👀 989 📊 8 ⚡