Skywork/Skywork-OR1-RL-Data

Skywork-OR1-RL-数据集是一个包含可验证的、具有挑战性的和多样化的数学问题（105K）和编码问题（14K）的数据集。该数据集用于训练Skywork-OR1模型系列，这是一个强大的数理和代码推理模型系列，通过大规模基于规则的强化学习以及精心设计的数据集和训练配方进行训练。数据集从开源数据集中选取、清洗和策划数学和编码问题，并对每个问题和模型进行模型感知难度估计，以及通过人工和LLM-as-a-Judge进行严格的质量评估，以确保训练效率和效果。

The Skywork-OR1-RL-Data dataset is a collection of verifiable, challenging, and diverse math problems (105K) and coding questions (14K). This dataset is used to train the Skywork-OR1 model series, which consists of powerful math and code reasoning models trained through large-scale rule-based reinforcement learning with carefully designed datasets and training recipes. The dataset selects, cleans, and curates math and coding problems from open-source datasets, conducting model-aware difficulty estimation for each problem and model, as well as rigorous quality assessment through both human and LLM-as-a-Judge to ensure training efficiency and effectiveness.