The paper proposes FERMI, a method that significantly improves membership inference attacks against tabular diffusion models by leveraging auxiliary relational information available during training, e…

The paper systematically characterizes column-level activation sparsity across various diffusion model architectures, demonstrating that element-level sparsity metrics significantly overestimate the a…

The paper introduces NaRA, a noise-aware LoRA technique that dynamically adapts fine-tuning parameters based on the noise level during diffusion, significantly improving the performance of Diffusion L…

The paper introduces DSL-LLaDA, a method that lightly adapts a pre-trained masked diffusion language model to perform continuous denoising in embedding space, significantly improving text generation q…

The paper introduces TabPrep, a feature engineering pipeline that systematically improves performance across various tabular machine learning models by addressing structural data patterns ignored by c…

The paper introduces Tail-risk Intrinsic Geometric Smoothing (TIGS), a plug-and-play, inference-time defense that suppresses backdoor attacks on LLMs by structurally smoothing the attention mechanism…

This paper investigates the phenomenon of 'copying' in Distribution Matching Distillation (DMD), finding that high-dimensional distillation causes student models to spontaneously reproduce the teacher…

The paper proposes Latent Geometric Chords (LGC) and LGC-H, a novel method that navigates decision boundaries using curvature-aware geometric search within a semantic manifold to generate high-fidelit…

While restricting a model to the theoretical Markov boundary can significantly improve prediction, the practical process of discovering and using this boundary is often computationally infeasible and…

The paper demonstrates that off-the-shelf image diffusion models, like Stable Diffusion, can be repurposed to generate synthetic structured data, posing a threat of ground truth drift in closed eviden…

TabChange proposes a novel framework to generate natural and minimally altered counterfactual instances in tabular data by precisely controlling attribute modifications based on their relationship str…

Dual-Guard introduces a dual-channel latent watermarking framework that simultaneously embeds global provenance and localized content anchors into diffusion images, achieving robust detection against…

The paper proposes FI-LDP-HGAT, a novel framework that combines a hierarchical graph attention network with feature-importance-aware anisotropic differential privacy to enable high-utility, privacy-pr…

The paper proposes using a mask-conditioned latent diffusion model to generate synthetic, labeled TEM images for data augmentation, achieving small but measurable performance improvements in defect de…

The paper introduces diffGHOST, a conditional diffusion model that generates synthetic, privacy-preserving mobility trajectories by explicitly mitigating sample memorization in the latent space.

The paper introduces a histogram-regularized latent diffusion model to synthesize highly realistic and subtype-specific pulmonary nodules in 3D CT volumes, addressing the limitations of existing metho…

The paper proposes VRPO, a reinforcement learning-based optimization strategy that replaces static alignment losses in diffusion models, significantly improving both convergence and image fidelity.

UMEDA introduces a novel graph federated learning framework that uses spectral signal processing and diffusion models to enable privacy-preserving, robust localization across clients with highly heter…