ADE：面向人类中心目标的智能体数据进化框架
原标题：ADE: Agentic Data Evolution Framework for Human-Centered Objectives
AI 摘要
arXiv 论文提出 Agentic Data Evolution (ADE) 框架，用于在弱验证条件下构建合成监督数据，通过观察-变异-选择闭环和稳态准入机制实现数据快照的持续改进。在 DEV300 上，ADE 将内在胜率从 50% 提升至 75.81%，外在胜率从 55.20% 提升至 68.86%，盲评专家对进化答案的偏好率达 66.11%。该框架在多种后训练方法、模型规模及任务上均表现出一致增益，资源已开源。
正文节选
tcboxmath \tl_set:Ne\tcbhighmathtcbhighmath ADE: Agentic Data Evolution Framework for Human-Centered Objectives Abstract Aligning large language models to human-centered objectives is difficult when targets are non-executable and context-dependent, limiting reliable verification and scalable supervision. Although synthetic data expands coverage, weak verification shifts the bottleneck from generation to selection. Noisy signals destabilize iterative refinement and can cause silent regressions. W