CENTAUR：在隐私保护 Transformer 推理中弥合隐私、效率与性能的不可能三角
CENTAUR: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference
AI总结：
本文提出融合随机置换与 SMPC 的 PPTI 框架 CENTAUR，通过适配 Transformer 结构的高效算法，在强隐私、明文级准确率和 5.0–30.4 倍加速之间取得平衡。
摘要：
随着 Transformers 等预训练模型在云平台上的部署日益增多，围绕模型参数和推理数据的隐私担忧正不断加剧。现有的隐私保护 Transformer 推理（PPTI）框架面临平衡隐私、效率与性能的“不可能三角”：基于安全多方计算（SMPC）的方法可确保强隐私性，但会产生高昂的计算开销并造成性能损失；相反，基于置换的方法能够实现接近明文的效率和准确率，却会因暴露敏感模型参数和中间结果而损害隐私。用单一方法弥合这一差距存在巨大挑战，因此本文提出 CENTAUR，这是一个开创性的 PPTI 框架，将随机置换与 SMPC 无缝集成，以应对“不可能三角”。通过设计适配 Transformer 模型结构特性的高效 PPTI 算法，CENTAUR 在隐私、效率和性能之间实现了前所未有的平衡。实验表明，CENTAUR 能够抵御多种数据重构攻击，达到明文级推理准确率，并将推理速度提升 5.0–30.4 倍，为安全高效的 AI 部署开辟了新的可能性。
英文摘要：
With the growing deployment of pre-trained models like Transformers on cloud platforms, privacy concerns about model parameters and inference data are intensifying. Existing Privacy-Preserving Transformer Inference (PPTI) frameworks face the "impossible trinity" of balancing privacy, efficiency, and performance: Secure Multi-Party Computation (SMPC)-based approaches ensure strong privacy but suffer from high computational overhead and performance losses; Conversely, permutation-based methods achieve near-plaintext efficiency and accuracy but compromise privacy by exposing sensitive model parameters and intermediate results. Bridging this gap with a single approach presents substantial challenges, motivating the introduction of CENTAUR, a groundbreaking PPTI framework that seamlessly integrates random permutations and SMPC to address the "impossible trinity". By designing efficient PPTI algorithms tailored to the structural properties of Transformer models, CENTAUR achieves an unprecedented balance among privacy, efficiency, and performance. Our experiments demonstrate CENTAUR's ability to resist diverse data reconstruction attacks, achieve plaintext-level inference accuracy, and boost inference speed by 5.0-30.4 times, unlocking new possibilities for secure and efficient AI deployment.
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。
