Diffusion Models

This paper proposes a one-shot channel surrogate called condition-wise Sinkhorn drifting for learned communication systems.

This paper introduces GEAR, a method for training a vector-quantized tokenizer and an autoregressive generator jointly and end-to-end, resolving the issue of non-differentiable VQ indices.

This paper investigates using uncertainty as a proxy for semantic correctness in diffusion-based medical image synthesis.

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.

This paper introduces MFlow, an adaptive parameterization for pixel-space diffusion models that learns to interpolate between two prediction methods, and ConvRefiner, a lightweight module to improve h…

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 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 proposes FLAME, a novel framework that detects AI-generated image forgeries by identifying intrinsic energy anomalies caused by the diffusion process, achieving state-of-the-art localization…

The paper analyzes that while multimodal large language models (MLLMs) offer superior semantic understanding for image generation, this enhanced capability significantly increases safety risks, partic…

The paper introduces Score-Guided Classification (SGC), a novel framework that uses an unsupervised anomaly score as a 'Pathological Prior' to guide EEG-based depression detection, overcoming the limi…

The paper introduces SHIFT, a training-free attack that exploits the vulnerability of diffusion-based watermarking by stochastically deflecting the generative trajectory, achieving high removal rates…

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…

This paper proposes a world model called FR3D that predicts a persistent 3D latent representation for future dynamic 3D reconstruction.

This paper proposes ReChannel, a method for dense prediction using a pretrained DiT model, which keeps the encoder but removes the decoder and adapts it with task LoRA. ReChannel maps each token to it…

This paper proposes Twins, a unified continuous token space for multimodal models using ViT and VAE features, and addresses optimization imbalance with a focal regression objective.

This paper introduces World Tracing, a method for generating complete 3D geometry from 2D images while preserving pixel alignment.

A new attack called BrushOff can remove robust watermarks from paintings by redrawing them with brush strokes.

The paper introduces Label-Augmented Conditional Diffusion (LACD), a method for learning complex conditional distributions using unlabeled data, and provides theoretical guarantees for its effectivene…

This paper presents RepFusion, a method that uses multimodal language models as noisy representation encoders to improve text-to-image synthesis.

This paper proposes a novel distillation framework for flow matching models that enables high-fidelity single-step generation.