arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.01757cs.CLcs.LG
收藏

Zero2Text: 无训练跨域反向攻击文本嵌入

Zero2Text: Zero-Training Cross-Domain Inversion Attacks on Textual Embeddings

School of Computing, Korea Advanced Institute of Science and Technology (KAIST)(韩国科学技术院计算学院);Graduate School of Information Security, Korea Advanced Institute of Science and Technology (KAIST)(韩国科学技术院信息安全研究生院);Department of Artificial Intelligence, University of Seoul, Seoul, Republic of Korea(首尔大学人工智能系);Department of Computer Science, University of Seoul, Seoul, Republic of Korea(首尔大学计算机科学系)

Doohyun Kim, Donghwa Kang, Kyungjae Lee, Hyeongboo Baek, Brent Byunghoon Kang

AI总结:

Zero2Text通过递归在线对齐机制,在无训练情况下实现跨域文本嵌入反向攻击,有效对抗传统防御手段,实验显示其在恢复句子方面表现优异。

摘要:

检索增强生成(RAG)的普及使向量数据库成为关键基础设施,但通过嵌入反向攻击引入了严重的隐私风险。现有方法面临根本性的权衡:基于优化的方法需要计算上不可行的查询,而基于对齐的方法则依赖于不现实的假设,即有可访问的领域内训练数据。这些限制使得它们在严格的黑盒和跨域设置中无效。为了打破这些障碍,我们引入了Zero2Text,一种基于递归在线对齐的新型无训练框架。与依赖静态数据集的方法不同,Zero2Text将LLM先验与动态岭回归机制结合,以迭代对齐生成到目标嵌入。我们进一步证明,标准防御,如差分隐私,无法有效缓解这种适应性威胁。在多样化的基准测试中进行了广泛的实验,验证了Zero2Text;值得注意的是,在MS MARCO对抗OpenAI受害模型时,其ROUGE-L和BLEU-2得分分别比基线高1.8倍和6.4倍,能够在没有单个泄露数据对的情况下从未知领域恢复句子。

英文摘要:

The proliferation of retrieval-augmented generation (RAG) has established vector databases as critical infrastructure, yet they introduce severe privacy risks via embedding inversion attacks. Existing paradigms face a fundamental trade-off: optimization-based methods require computationally prohibitive queries, while alignment-based approaches hinge on the unrealistic assumption of accessible in-domain training data. These constraints render them ineffective in strict black-box and cross-domain settings. To dismantle these barriers, we introduce Zero2Text, a novel training-free framework based on recursive online alignment. Unlike methods relying on static datasets, Zero2Text synergizes LLM priors with a dynamic ridge regression mechanism to iteratively align generation to the target embedding on-the-fly. We further demonstrate that standard defenses, such as differential privacy, fail to effectively mitigate this adaptive threat. Extensive experiments across diverse benchmarks validate Zero2Text; notably, on MS MARCO against the OpenAI victim model, it achieves 1.8x higher ROUGE-L and 6.4x higher BLEU-2 scores compared to baselines, recovering sentences from unknown domains without a single leaked data pair.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑