As modern text-to-image (T2I) models draw closer to synthesizing highly realistic content, the threat of unsafe content generation grows, and it becomes paramount to exercise control. Existing approaches steer these models by applying Euclidean adjustments to text embeddings, redirecting the generation away from unsafe concepts. In this work, we introduce hyperbolic control (HyCon): a novel control mechanism based on parallel transport that leverages semantically aligned hyperbolic representation space to yield more expressive and stable manipulation of concepts. HyCon reuses off-the-shelf generative models and a state-of-the-art hyperbolic text encoder, linked via a lightweight adapter. HyCon achieves state-of-the-art results across four safety benchmarks and four T2I backbones, showing that hyperbolic steering is a practical and flexible approach for more reliable T2I generation.
Latent Variable Modelling with Hyperbolic Normalizing Flows
McGill University(麦吉尔大学);Mila(米拉研究所);University of Toronto(多伦多大学);Vector Institute(向量研究所)
近似后验分布的选择在随机变分推理(SVI)中起着核心作用。一个有效的解决方案是使用归一化流\cut{defined on Euclidean spaces}来构造灵活的后验分布。然而,现有的规范化流的一个关键限制是,它们被限制在欧几里德空间,并且不适合对具有底层层次结构的数据进行建模。为了解决这个基本的限制,我们提出了规范化流的双曲空间的第一个扩展。我们首先提升规范化流双曲空间使用耦合变换定义的切丛,称为切线耦合($\mathcal{TC}$)。我们进一步介绍了包裹双曲面耦合($\mathcal{W}\mathbb{H}C$),一个完全可逆的和可学习的变换,明确利用双曲空间的几何结构,允许表达后验,同时是有效的采样。我们证明了我们的新的归一化流的双曲VAE和欧几里得归一化流的功效。我们的方法提高了密度估计的性能,以及表现出分层结构的真实世界图数据的重建。最后,我们表明,我们的方法可以用来权力的生成模型在分层数据使用双曲型潜变量。
FLAT: Revealing Hidden Latent-Conditioned Backdoor Failures in Federated Learning
College of Engineering and Computer Science, VinUniversity(VinUniversity工程与计算机科学学院);VinUni-Illinois Smart Health Center, VinUniversity(VinUniversity-VinUni-Illinois智能健康中心);Siebel School of Computing and Data Science, University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校Siebel计算与数据科学学院);Faculty of Computer Science and Information Technology, Universiti Malaya(马来亚大学计算机科学与信息技术学院)