揭示流匹配模型中训练数据的隐蔽记忆信号，可用来识别训练数据成员。
Where Flow Matching Leaks: Characterising Membership Signals Along the Interpolation Path
- 通过线性插值路径分析流匹配模型的记忆泄露机制。
- 发现重建误差随插值参数λ呈钟形曲线，峰值位置可解析推导。
- 首次实现基于λ解析结构的成员推理攻击，适用于隐私安全研究者。
理解生成模型中的记忆现象仍具挑战性，对版权和隐私有重要影响。除了直接复现外，模型可能编码训练数据的细微痕迹，这些痕迹虽不体现在输出中，却仍可被探测。我们称此类可测量的不对称性为 extit{成员信号}，并研究其在流匹配（Flow Matching）中的表现，该方法广泛应用于部署的生成系统。我们分析定义标准流匹配训练的线性插值路径 $X_λ= (1-λ)X_0 + λX_1$。结果表明，训练数据与测试数据的重建差异在λ上呈现钟形曲线，该差异在训练过程中累积，而验证指标保持稳定。在高斯假设下，我们推导出峰值位置的闭式解。我们在音频和图像数据集上验证了这一钟形结构的普适性，且当假设成立时，峰值预测准确。作为概念验证，我们利用该λ解析结构实施成员推理攻击，成功区分训练集成员与非成员。
原文摘要 · Abstract (English)
Understanding memorization in generative models remains challenging, with implications for copyright and privacy. Beyond verbatim reproduction, models can encode subtler traces of their training data that never surface in their outputs yet remain exploitable. We refer to these measurable asymmetries as the \emph{membership signal}, and we study this regime for Flow Matching, which are increasingly used in deployed generative systems. We analyze the linear interpolation path $X_λ= (1-λ)X_0 + λX_1$ that defines standard Flow Matching training. We show that a gap exists between the reconstruction of train and test data that follows a bell-shaped curve over $λ$, which accumulates during training, while the validation metrics remain stable. The signal has a maximum whose location we derive in closed form under Gaussian assumptions. We validate these predictions on both audio and images and show that the bell-shaped structure is universal, while the peak prediction holds when our assumptions are satisfied. As a proof of concept, we exploit this specific $λ$-resolved structure to perform a Membership Inference Attack, distinguishing members of the training set from non-members.
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。