nvidia/Nemotron-RL-knowledge-mcqa

Nemotron-RL-knowledge-mcqa是一个多领域的合成多项选择问答（MCQA）数据集，包含基于知识的问题。它结合并优化了OpenScienceReasoning-2数据集的子集以及其他非结构化来源（如书籍和文章）。该数据集使用Qwen3-32B、Qwen3-235B-A22B-Instruct-2507和DeepSeek-R1-0528模型创建。每个样本包含一个问题和多个答案选项，其中一个为正确答案。数据集涵盖广泛的领域，包括物理、生物、化学、数学、计算机科学、工程、人文、法律等。该数据集作为NVIDIA NeMo Gym的一部分发布，用于训练大型语言模型的强化学习环境。数据集可用于商业用途，采用CC BY 4.0许可证。

The Nemotron-RL-knowledge-mcqa is a multi-domain synthetic multiple-choice question-answering (MCQA) dataset containing knowledge based questions. It combines and refines subsets of the OpenScienceReasoning-2 dataset and other unstructured sources such as books and articles. The dataset was created using Qwen3-32B, Qwen3-235B-A22B-Instruct-2507, and DeepSeek-R1-0528. Each sample consists of a question with multiple answer options and one correct answer. The dataset spans a broad range of domains, including physics, biology, chemistry, mathematics, computer science, engineering, humanities, law, and others. This dataset is released as part of NVIDIA NeMo Gym, a framework for building reinforcement learning environments to train large language models. The dataset is ready for commercial use and is released under the CC BY 4.0 license.