面向模型专业化的自主智能体数据工程探索

摘要

大语言模型（LLMs）在通用任务上展现出强大性能，但常因缺乏高质量领域特定数据而难以适配专业领域。现有基于LLM的数据策管方法主要依赖人工设计的工作流程，尚未验证LLM能否自主执行端到端的数据工程流水线以实现模型专业化。我们正式提出"自主智能体数据工程"这一新任务，旨在评估LLM作为自主数据工程师，通过端到端数据策管驱动模型专业化的能力。我们将数据视为可优化组件，研究智能体如何规划、生成并迭代优化多领域训练数据，并以训练后性能提升为导向。实验表明，自主LLM数据工程师带来了显著增益：GPT-5.2构建的训练课程使学生模型性能提升了57.29%，完全通过基于智能体的迭代数据适应实现。通过揭示潜力与瓶颈，本研究将自主数据工程确立为一项可量化的能力，并为智能体驱动的模型专业化指明方向。代码将发布于https://github.com/zjunlp/DataAgent。

English

Large Language Models (LLMs) have demonstrated strong performance on general tasks, while often struggling to adapt to specialized domains without high-quality domain-specific data. Existing LLM-based data curation methods primarily rely on human-designed workflows, leaving it unexamined whether LLMs can autonomously execute an end-to-end data engineering pipeline for model specialization. We formalize Autonomous Agentic Data Engineering, a novel task designed to evaluate LLMs as autonomous data engineers that drive model specialization through end-to-end data curation. We frame data as an optimizable component and study agents that plan, generate, and iteratively optimize training data across multiple domains, guided by post-training performance improvement. Experiments show that autonomous LLM data engineers yield substantial gains, as GPT-5.2 constructs a training curriculum that improves a student model by 57.29\%, entirely through iterative, agent-driven data adaptation. By illuminating both potential and bottlenecks, our study establishes autonomous data engineering as a measurable capability and charts a path toward agent-driven model specializationCode will be released at https://github.com/zjunlp/DataAgent..