ACIDE/user-vlm-bias-bench
收藏Hugging Face2025-02-14 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/ACIDE/user-vlm-bias-bench
下载链接
链接失效反馈官方服务:
资源简介:
VLM Benchmark for Bias Mitigation数据集用于评估模型在选择BiasVision-DPO数据集中的100个条目时的道德价值观和不偏不倚性。数据集包含字段:prompt、chosen、rejected、image和category,划分为训练集,共有100个示例。
The VLM Benchmark for Bias Mitigation dataset is used to assess the models moral values and impartiality in selecting 100 entries from the BiasVision-DPO dataset. The dataset includes fields: prompt, chosen, rejected, image, and category, and is split into a training set with a total of 100 examples.
提供机构:
ACIDE



