McGill-NLP/AdvBench-IR
收藏Hugging Face2025-03-12 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/McGill-NLP/AdvBench-IR
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含针对AdvBench查询生成的恶意文档。这些文档使用Mistral-7B-Instruct-v0.2语言模型创建,用于问题回答和文本到文本生成的任务。
This dataset includes malicious documents generated in response to AdvBench queries using the Mistral-7B-Instruct-v0.2 language model for question-answering and text-to-text generation tasks.
提供机构:
McGill-NLP



