Model stealing attacks pose an existential threat to Machine Learning as a Service (MLaaS), allowing adversaries to replicate proprietary models for a fraction of their training cost. While Data-Free Model Extraction (DFME) has emerged as a stealthy vector, it remains fundamentally constrained by the "Cold Start" problem: GAN-based adversaries waste thousands of queries converging from random noise to meaningful data. We propose DiMEx, a framework that weaponizes the rich semantic priors of pre-trained Latent Diffusion Models to bypass this initialization barrier entirely. By employing Random Embedding Bayesian Optimization (REMBO) within the generator's latent space, DiMEx synthesizes high-fidelity queries immediately, achieving 52.1 percent agreement on SVHN with just 2,000 queries - outperforming state-of-the-art GAN baselines by over 16 percent. To counter this highly semantic threat, we introduce the Hybrid Stateful Ensemble (HSE) defense, which identifies the unique "optimization trajectory" of latent-space attacks. Our results demonstrate that while DiMEx evades static distribution detectors, HSE exploits this temporal signature to suppress attack success rates to 21.6 percent with negligible latency.

翻译：模型窃取攻击对机器学习即服务（MLaaS）构成生存性威胁，使攻击者能够以远低于训练成本的代价复制专有模型。尽管无数据模型提取（DFME）已成为一种隐蔽的攻击途径，但其本质上仍受限于"冷启动"问题：基于生成对抗网络（GAN）的攻击者需要耗费数千次查询才能从随机噪声收敛至有效数据。本文提出DiMEx框架，该框架利用预训练潜在扩散模型丰富的语义先验知识，完全绕过该初始化壁垒。通过在生成器的潜在空间中采用随机嵌入贝叶斯优化（REMBO），DiMEx能够即时合成高保真查询，仅用2,000次查询即在SVHN数据集上达到52.1%的模型一致性——超越当前最先进的GAN基线方法超过16个百分点。为应对这种高语义威胁，我们提出混合状态集成（HSE）防御机制，该机制通过识别潜在空间攻击特有的"优化轨迹"实现检测。实验结果表明，虽然DiMEx能够规避静态分布检测器，但HSE利用其时序特征将攻击成功率压制至21.6%，且延迟可忽略不计。