用多智能体框架生成可控教育人格，支持100万学生数据集。
HACHIMI: Scalable and Controllable Student Persona Generation via Orchestrated Agents
- 构建多智能体框架，分三步生成符合教育理论的人格。
- 生成100万学生人格，各年级配额精准，语义去重降低重复。
- 适合教育仿真与群体评测，尤其关注数学与求知欲的拟合度。
学生人格（SPs）正成为教育大模型的基础架构，但以往方法依赖随意提示或手工设计，难以控制教育理论与人群分布。本文将此问题形式化为理论对齐与分布可控的人格生成（TAD-PG），提出HACHIMI多智能体框架，通过提议-验证-修正流程生成符合理论、满足配额的人格。该框架将每类人格分解为理论锚定的教育结构，利用神经符号验证器施加发展与心理约束，并结合分层采样与语义去重减少模式崩溃。最终产出包含100万条人格的HACHIMI-1M数据集，覆盖1至12年级。内在评估显示结构有效性接近完美，配额准确且多样性高；外在评估中，人格作为学生代理回答CEPS与PISA 2022问卷，在16个组别中，数学能力与好奇心/成长维度与真人高度一致，而课堂氛围与幸福感维度仅中等一致，揭示出拟合度梯度。所有人格均基于Qwen2.5-72B生成，为群体级基准测试与社会科学模拟提供标准化合成学生群体。资源见https://github.com/ZeroLoss-Lab/HACHIMI。
原文摘要 · Abstract (English)
Student Personas (SPs) are emerging as infrastructure for educational LLMs, yet prior work often relies on ad-hoc prompting or hand-crafted profiles with limited control over educational theory and population distributions. We formalize this as Theory-Aligned and Distribution-Controllable Persona Generation (TAD-PG) and introduce HACHIMI, a multi-agent Propose-Validate-Revise framework that generates theory-aligned, quota-controlled personas. HACHIMI factorizes each persona into a theory-anchored educational schema, enforces developmental and psychological constraints via a neuro-symbolic validator, and combines stratified sampling with semantic deduplication to reduce mode collapse. The resulting HACHIMI-1M corpus comprises 1 million personas for Grades 1-12. Intrinsic evaluation shows near-perfect schema validity, accurate quotas, and substantial diversity, while external evaluation instantiates personas as student agents answering CEPS and PISA 2022 surveys; across 16 cohorts, math and curiosity/growth constructs align strongly between humans and agents, whereas classroom-climate and well-being constructs are only moderately aligned, revealing a fidelity gradient. All personas are generated with Qwen2.5-72B, and HACHIMI provides a standardized synthetic student population for group-level benchmarking and social-science simulations. Resources available at https://github.com/ZeroLoss-Lab/HACHIMI
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。