datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
aime24_figuresThe 30 problems from AIME 2024 with all ASY code for figures.
Citation Information
@misc{muennighoff2025s1simpletesttimescaling,
title={s1: Simple test-time scaling},
author={Niklas Muennighoff and Zitong Yang and Weijia Shi and Xiang Lisa Li and Li Fei-Fei and Hannaneh Hajishirzi and Luke Zettlemoyer and Percy Liang and Emmanuel Candès and Tatsunori Hashimoto},
year={2025},
eprint={2501.19393},
archivePrefix={arXiv},
primaryClass={cs.CL}… See the full description on the dataset page: https://huggingface.co/datasets/simplescaling/aime24_figures.aime_figuresThe 90 problems from AIME 2022, 2023, 2024 with all ASY code for figures.
Citation Information
@misc{muennighoff2025s1simpletesttimescaling,
title={s1: Simple test-time scaling},
author={Niklas Muennighoff and Zitong Yang and Weijia Shi and Xiang Lisa Li and Li Fei-Fei and Hannaneh Hajishirzi and Luke Zettlemoyer and Percy Liang and Emmanuel Candès and Tatsunori Hashimoto},
year={2025},
eprint={2501.19393},
archivePrefix={arXiv}… See the full description on the dataset page: https://huggingface.co/datasets/simplescaling/aime_figures.figure4-visualization-dataset
Ordinal-Reference Visualization Dataset
Created: 2026-07-24 UTCStatus: final frozen 50-case package
This package is a hand-curated qualitative dataset for horizontal, same-case image-editing comparisons. Each image contains repeated instances of one subject class arranged in one or two visually legible horizontal rows. The cases deliberately span food, plants, household and laboratory objects, toys, vehicles, wildlife, and people so that ordinal grounding is tested across more… See the full description on the dataset page: https://huggingface.co/datasets/XIAOHAOYANG/figure4-visualization-dataset.published-houseplant-light-figuresChinese_Historical_Figures_Dialogue
华夏群雄历史人物对话数据集 (Chinese Historical Figures Dialogue Dataset)
数据集概述
华夏群雄历史人物对话数据集是一个大规模、高质量的中文历史人物对话数据集，专为训练和评估中文大语言模型的历史人物扮演能力而设计。该数据集通过AI大模型与中国历史上重要人物的虚拟对话生成，涵盖了从上古时代到清朝的150位历史人物。
数据集特色
🎯 核心优势
人物多样性：涵盖150位中国历史重要人物，包括政治家、文学家、哲学家、军事家等
内容均衡性：每个人物最多700行内容，确保数据分布均衡，避免过度集中
对话真实性：严格保持历史人物的第一人称视角和时代背景
任务丰富性：包含6种不同类型的对话任务
📊 数据规模
历史人物：150 位（从大禹到清朝）
对话样本：100,000+ 条高质量对话
任务类型：6种核心对话类型
🏗️ 数据结构
数据集采用Alpaca格式，包含以下字段：
{
"instruction":… See the full description on the dataset page: https://huggingface.co/datasets/QingYuYunTu/Chinese_Historical_Figures_Dialogue.aime25_figures