RECAP-Project/EchoTrace
收藏Hugging Face2025-11-16 更新2025-11-30 收录
下载链接:
https://hf-mirror.com/datasets/RECAP-Project/EchoTrace
下载链接
链接失效反馈官方服务:
资源简介:
EchoTrace数据集是一个用于评估和分析和大型语言模型(LLM)中的记忆化和训练数据泄露的基准。该数据集包含35本叙事书籍,分为公共领域书籍、版权畅销书和非训练书籍。每个书籍段落都包含高级摘要、原文文本段和事件级元数据。
The EchoTrace dataset is a benchmark designed to evaluate and analyze memorization and training data exposure in Large Language Models (LLMs). It consists of 35 narrative books, divided into public domain books, copyrighted bestsellers, and non-training books. Each book segment includes a high-level summary, the verbatim text segment, and event-level metadata.
提供机构:
RECAP-Project
搜集汇总
数据集介绍

以上内容由遇见数据集搜集并总结生成



