LifelongAlignment/aifgen-short-piecewise
收藏Hugging Face2025-05-16 更新2025-11-01 收录
下载链接：
https://hf-mirror.com/datasets/LifelongAlignment/aifgen-short-piecewise
下载链接
链接失效反馈官方服务：
资源简介：
这是一个为终身强化学习适应性训练而创建的连续数据集，包含两个任务：数学、社会科学、物理或化学的非平凡问题，一个是提示性回答任务，另一个是直接回答任务。数据集由Mila - Quebec AI Institute和Complex Data Lab团队使用AIF-Gen框架生成，适用于大型语言模型的微调。
This is a continuous dataset created for lifelong reinforcement learning adaptation training, containing two tasks: one is a hinted answer task for non-trivial math, social science, physics, or chemistry questions, and the other is a direct answer task. The dataset is generated using the AIF-Gen framework by the Mila - Quebec AI Institute and the Complex Data Lab, suitable for fine-tuning large language models.
提供机构：
LifelongAlignment
