提出可高效加密序列推理的同态状态空间模型，兼顾安全与速度。
Public-Decay Homomorphic State Space Models for Private Sequence Inference
- 用公开衰减机制更新加密状态，避免复杂乘法运算。
- 在影评和情感数据集上准确率达0.7505和0.7420，媲美明文模型。
- 相比同类方法快5倍，内存占用更低，适合实际部署。
全同态加密（FHE）改变了序列模型设计，旋转、加密乘法、密文生成、乘法深度和刷新操作带来的开销远超普通神经网络成本。本文提出公开衰减同态状态空间模型（HSSM），其递归/状态空间模块通过密文-明文公开衰减更新携带状态，而密文-密文乘法则保持在本地写路径。该设计使加密状态在整个序列中保持固定。评估流程将客户端的分词、冻结fastText查找、投影、截断、加密、解密和阈值处理与服务端基于受限投影特征的加密计算分离。在完整的Rotten Tomatoes和SST-2验证集上，加密HSSM路径精确匹配明文分类结果，准确率分别为0.7505和0.7420。相较于对HE友好的多项式注意力，在相同fastText任务上，HSSM在全序列任务质量上持平或超越，且运行速度约快5倍。配对的L40S操作级对比显示，其延迟比缓存最终标记的多项式注意力低1.34–1.62倍，比全序列多项式注意力低30–258倍，且加密状态逻辑占用更小。在T=16/32的比较中，加密公开线性输入与Q/K/V投影下，投影的HSSM可在深度8/环大小32768下成功运行，而投影注意力需深度10/环大小65536。在相同的T=8 OpenFHE/FIDESlib追踪中，两者均在最终层第3层、噪声尺度度为2处完成。这些结果表明，公开衰减是一种实用的FHE协同设计杠杆，适用于基于受限投影特征的加密序列推理。
原文摘要 · Abstract (English)
Fully homomorphic encryption (FHE) changes sequence-model design because rotations, encrypted products, ciphertext materialization, multiplicative depth, and bootstrapping pressure can dominate ordinary neural-network costs. This paper presents public-decay homomorphic state space models (HSSMs), recurrent/state-space blocks whose carried state is updated through ciphertext-plaintext public decay while ciphertext-ciphertext multiplication remains on a local write path. The design keeps a fixed encrypted state across the sequence. The evaluated workflow separates client-side tokenization, frozen fastText lookup, projection, clipping, encryption, decryption, and thresholding from server-side encrypted evaluation over bounded projected features. On full Rotten Tomatoes and SST-2 validation splits, the encrypted HSSM path exactly matches plaintext classifications and reaches 0.7505 and 0.7420 accuracy. Against HE-friendly polynomial attention on the same fastText workloads, HSSM matches or exceeds full-sequence task quality while running about 5x faster. Paired L40S operation-level rows show 1.34-1.62x lower latency than cached final-token polynomial attention, 30-258x lower latency than full-sequence polynomial attention, and lower logical encrypted-state footprint. A T = 16/32 comparator with encrypted public-linear input and Q/K/V projections shows projected HSSM succeeding under depth 8/ring 32768, while projected attention succeeds under depth 10/ring 65536. A matched T = 8 OpenFHE/FIDESlib trace finishes at final level 3 and noise-scale degree 2 on both backends. These results make public-decay carry a practical FHE co-design lever for encrypted sequence inference from bounded projected features.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。