#image synthesis

9 results
cs.CR2026

AnchorMark: Robust Diffusion Watermarking via Latent-Space Rotation Synchrony

Yuqi Qian, Yun Cao, Haocheng Fu +3

The paper proposes AnchorMark, a training‑free inversion‑based watermarking method for diffusion models that uses a latent‑space rotation synchrony property to embed a central anch…

#watermarking#diffusion models#latent space#rotation robustness
cs.LG2026

Amortized Moment Matching for Visual Generation

Wenze Liu, Xintao Wang, Pengfei Wan +1

The paper introduces amortized moment matching, using neural networks to learn data moments as training signals, and proposes the Amortized Fréchet Distance loss to improve one-ste…

#generative models#moment matching#diffusion models#image synthesis
cs.LG2026

Flow Map Learning via Nongradient Vector Flow

Mark Goldstein, Anshuk Uppal, Raghav Singhal +2

The paper proposes SGFlow, a method that learns flow maps for diffusion and flow‑based generative models without requiring model invertibility or backpropagation through repeated m…

#generative modeling#diffusion models#flow-based models#ode inference
cs.CV2026

Do Unified Multimodal Models Think in One Space? A Lens Through Cross-Branch Steering

Yu Wang, Sharon Li

The paper investigates whether unified multimodal models share a common semantic space by introducing cross-branch semantic steering, showing that semantic directions from the unde…

#unified multimodal models#semantic steering#cross-modal transfer#image synthesis
cs.CV2026

LumaGuide: Distribution Shaping for Training-Free HDR Generation in Diffusion Models

Bowen Chen, Shreshth Saini, Balu Adsumilli +1

LumaGuide is a training‑free framework that steers the sampling of pretrained diffusion models by matching target luminance distributions, enabling high dynamic range (HDR) image g…

#diffusion models#hdr imaging#distribution shaping#energy-based guidance
cs.CV2026

Controllable Generation of Diverse Dermatological Imagery for Fair and Efficient Malignancy Classification

Héctor Carrión, Narges Norouzi

The paper presents cgDDI, a controllable generative framework that creates realistic, diverse dermatological images—including rare lesions on varied skin tones—to improve malignanc…

#dermatology#synthetic data generation#fairness#image synthesis
cs.CV2026

Conditioning Residuals for Diffusion Models via Representation Feedback

Weilai Xiang, Hongyu Yang, Di Huang +1

The paper introduces Conditioning Residuals, a lightweight feedback mechanism that injects compact feature summaries back into the conditioning embeddings of diffusion model backbo…

#diffusion models#conditional generation#representation feedback#UNet
cs.CV2026

The Devil Is in the Leakage: A Disentangled Dual-Purification Framework for High-Fidelity Hairstyle Transfer

Jijie Li, Jiankuo Zhao, Xiangyu Zhu +1

The paper introduces a Dual‑Purification Framework that reduces identity and geometric leakage in diffusion‑based hairstyle transfer, enabling high‑fidelity, identity‑preserving po…

#hairstyle transfer#image synthesis#generative diffusion models#identity preservation
cs.CV2026

Self-Consistent Flow: Unifying Velocity and Endpoint Prediction for Rectified Flow Models

Xu Han, Jiajing Hu, Li-Ping Liu

The paper introduces Self-Consistent Flow (SC-Flow), a method that jointly trains a single network to predict both local velocity and data endpoint in rectified-flow generative mod…

#generative modeling#rectified flow#velocity prediction#endpoint prediction