用单次生成实现高效可靠的教育内容自动化，支持幻灯片与互动网页。
CogEvol: Towards Efficient and Reliable Learning Environment Generation
- 输入课程简述即可一键生成结构化幻灯片或可运行的交互式网页。
- 生成速度达17秒/幻灯片、59秒/网页，参数量仅为主流编码模型的1/26.9。
- 开源4B版本，兼容国产芯片，适合大规模智能教育系统部署。
我们提出CogEvol，一套专为学习环境生成设计的模型家族：仅需一次处理，即可将课程简述转化为结构化JSON幻灯片或自包含的交互式HTML页面。在22万次生产请求中，幻灯片平均生成时间仅17秒，交互页为59秒，取代了以往耗时数分钟的多轮代理协作流程。可靠性通过真实失败案例构建的53,687个验证过的SFT样本实现，结合规则+视觉语言模型（VLM）奖励驱动的GRPO强化学习，修复了一次奖励黑客导致的不可玩游戏问题。CogEvol-27B在幻灯片质量评测中得分为83.7，在500项交互式网页基准测试中得分为63.7，参数量仅为旗舰编码模型的26.9倍。与OpenMAIC团队合作，已投入实际生产流量。开源的CogEvol-4B模型在Apache 2.0许可下发布于GitHub，外部旗舰模型均在相同测试套件和统一框架下评估。支架编辑进一步降低交互页生成成本约76%，全栈可在国产Ascend加速器上运行，性能媲美A800 GPU，显著降低规模化AI教育的单位成本。
原文摘要 · Abstract (English)
We present CogEvol, a family of models trained specifically for Learning Environment Generation: turning a course brief into a finished learning artifact (structured-JSON slides or self-contained interactive HTML pages) in a single pass. Across 220k production requests, CogEvol completes a slide in a median of 17 seconds and an interactive page in 59, replacing minutes-long multi-turn agent scaffolding. Reliability is enforced rather than hoped for: a production-grounded data pipeline turns real failures into 53,687 verified SFT samples, and a hybrid rule-plus-VLM reward drives GRPO-based RL, hardened after we caught and fixed a reward-hacking episode that produced visually convincing but unplayable games. CogEvol-27B scores 83.7 on slide quality and 63.7 on a 500-case interactive-HTML benchmark with 26.9x fewer parameters than flagship coding models, and, in collaboration with the OpenMAIC team, serves their live production traffic. CogEvol-4B is released openly under the Apache 2.0 license at https://github.com/CogEvol/CogEvol-4B; external flagships are measured on the same suites under the identical harness. Scaffold editing cuts interactive-page generation cost by a further ~76%, and the full stack runs on domestic Ascend accelerators at application-level parity with A800 GPUs, lowering the unit cost of AI-native education at scale.
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。