相关数据集
SIG
SIG数据集是由作者团队创建的，用于评估检索增强语言模型在处理具有伪迹特征的数据时的鲁棒性。该数据集通过SURE框架自动注入伪迹特征，并确保因果特征保持不变，从而生成具有挑战性的测试实例。数据集来源于NQ-Open查询和维基百科文档，旨在解决检索增强生成系统在现实世界应用中的鲁棒性问题。
arXiv2025-03-08 更新390
autoevaluate/autoeval-staging-eval-project-00ac2adb-9115199
--- type: predictions tags: - autotrain - evaluation datasets: - cifar10 eval_info: task: image_multi_class_classification model: karthiksv/vit-base-patch16-224-cifar10 metrics: [] dataset_nam
Hugging Face2022-06-29 更新50
Comparisons of the annotation accuracy rates for each category of the PASCAL VOC 2012 dataset based on different algorithms after 100 training iterations.
Comparisons of the annotation accuracy rates for each category of the PASCAL VOC 2012 dataset based on different algorithms after 100 training iterations.
Figshare2020-06-01 更新20
tunnel-28.zip
Tunnel-28 Core Dataset: This dataset provides original tunnel images paired with precise pixel-level segmentation maps, generated using LABELMA. The annotations focus on three critical regions:
Figshare2025-11-06 更新50
LiDAR Degeneration Test Dataset (Flatwall experiment)
A dataset for evaluating LiDAR-IMU SLAM performance in situations where LiDAR data becomes completely degenerated. A Livox Avia was used to record point cloud and IMU data, and the LiDAR was moved whi
NIAID Data Ecosystem20
