arXiv · cs.AI · Sruly Rosenblat, Tim O'Reilly, Ilan Strauss · 10 小时前模型评分53
研究用成员推断攻击检测 OpenAI 模型对未授权非公开书籍的识别
Beyond Public Access in LLM Pre-Training Data
导读
研究团队用 34 本 O'Reilly Media 版权书籍（13,962 个段落）构建数据集，通过 DE-COP 成员推断攻击测试 OpenAI 模型是否识别出未获授权训练的非公开付费内容。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
事件类型：研究发布
证据摘录
- “Using a legally obtained dataset of 34 copyrighted O'Reilly Media books (13,962 paragraphs), we apply the DE-COP membership inference attack to test whether OpenAI models show recognition of non-public (pay-walled) content they were not licensed to train on.”
- “Results remain preliminary given the small book sample, possible contamination of post-cutoff controls, and differences across models in size, architecture, and training composition.”
来源：arXiv · cs.AI · arxiv.org