SafeStyle is a training‑free framework that injects calibrated style residuals into frozen diffusion models for reference‑guided stylization. It estimates style‑supported and content‑associated subspaces from small calibration sets, then transports purified style evidence across adaptive spatial granularity while limiting its influence with a residual‑norm budget. Experiments on texture‑ and geometry‑dominant styles show high DINO style similarity (0.432–0.474) with minimal semantic leakage (0.8%).
By Zhangping Yang, Min Li, Song Yan, Rong Gao, Xinliang Bi, Guanye Xiong, Yujie He
The paper introduces SuRe, a framework that mitigates semantic leakage in image-guided zero-shot style transfer by regulating key–value coupling in diffusion model self‑attention. SuRe disrupts the native correspondence between style keys and values through a value shuffle technique, then progressively recouples them to preserve style fidelity while improving content similarity. Experiments on SD1.5, SDXL, SD3, and FLUX.1‑dev show that SuRe outperforms 18 existing methods in maintaining style similarity and consistently enhancing content similarity, with natural generalization to multi‑style transfer.
By Haojun Tang, Qiwei Lin, Tongda Xu, Wenda Zhao, Yian Zhao, Yu Zhang, Lida Huang, Yan Wang
arXiv:2608. 19719v1 Announce Type: cross Abstract: Reference-based diffusion stylization requires separating target geometry from transferable appearance.
By Jingtao Zhang, Haorui Gao, Youqing Liang, Zeming Liu
arXiv:2602.08059v2 Announce Type: replace-cross
Abstract: Text-to-image diffusion models can reproduce specific artists visual styles at extremely low cost, raising copyright and deployment safety co...
By Tong Zhang, Ru Zhang, Jianyi Liu
Reference-based diffusion stylization requires separating target geometry from transferable appearance. Existing tuning-based methods often rely on aligned content-style-target triplets or auxiliary visual encoders, which increases data cost and can transfer unintended scene structure from the style reference.
arXiv:2605.23891v2 Announce Type: replace
Abstract: Mask-free video object insertion has emerged as a challenging task, requiring harmonious integration of reference objects into source videos. Howev...
By Xiao Cao, Yansong Qu, Xiangzhen Chang, Wen Xiao, Jiakui Hu, Heyuan Li, Jialun Liu, Zhiyong Huang, Xuelong Li
DiDE is a training‑free framework that enables independent control of color and texture when stylizing 3D assets generated by rectified flow‑based image‑to‑3D models. It leverages the overcomplete latent space of these models, partitioning channels so that texture occupies a small subset while color is encoded in a free subspace. Experiments on the newly collected Disen3D‑Bench benchmark demonstrate that DiDE outperforms existing 2D and 3D stylization methods in color fidelity, texture transfer, and content preservation.
By Tao Wu, Alexandra Gomez-Villa, Senmao Li, Yaxing Wang, Joost van de Weijer, Kai Wang
Abstract‑LoRA introduces a lightweight LoRA training approach that targets specific U‑Net blocks in diffusion models to improve single‑image style transfer. By refining block selection, adding more blocks, and using clustering‑based style abstraction, it better disentangles and balances style and content compared to prior methods like B‑LoRA. Experiments show that the method produces more harmonious artistic images while quantitatively preserving both style and content.
By Xinglin Hu
GraftSR is a diffusion-based super‑resolution framework that uses reference images of the same object to guide texture restoration, mitigating hallucination. It introduces a dual‑mask reference guidance mechanism to decouple texture extraction from application, avoiding reliance on spatial alignment. The authors also release TexRefSR‑141K, a large dataset of reference pairs with spatial masks, and show that GraftSR outperforms existing methods on the TexRefSR‑Eval benchmark, reducing LPIPS by 20.2%.
By Qifan Yu, Haoran Bai, Zongyao He, Weijie He, Sibin Deng, Honggang Qi, Ying Chen
MAST (Mask‑Guided Attention Control for Training‑Free Regional‑Multi Style Transfer) is a framework that enables diffusion models to apply multiple reference styles to user‑specified regions of a content image without any training or optimization. It introduces logit‑level attention mass allocation, sharpness‑aware temperature scaling, and discrepancy‑aware detail injection to address mass allocation, selectivity, and detail loss problems in regional‑multi style transfer. Experiments with two to five styles show that MAST outperforms baselines in ArtFID, FID, and R‑FID, achieving high regional style fidelity, content preservation, and scalability.
By Dongkyung Kang, Jaeyeon Hwang, Junseo Park, Minji Kang, Yeryeong Lee, Beomseok Ko, Hanyoung Roh, Jeongmin Shin, Hyeryung Jang
Artistic image synthesis aims to recreate the expressive visual identity of a target artist, yet existing methods often fail to capture an artist's global style. Conventional style transfer methods transfer the style of one or a few reference artworks to a content image in a One-to-One manner, making them effective for artwork-level stylization but limited in representing the broader stylistic distribution of an artist.
RASteer is a training‑free method for concept erasure in text‑to‑image diffusion models that retains other concepts. It constructs a retain subspace from concepts to preserve, then removes components aligned with this subspace from the erasure direction using Retain‑Orthogonal Steering (ROS). Overlap‑Adaptive Calibration (OAC) further adjusts the removal of shared components at each layer and denoising step, balancing target erasure with concept preservation. Experiments show RASteer matches or outperforms existing activation steering and weight editing baselines on unsafe‑content, instance, and artistic‑style erasure tasks across multiple backbones and benchmarks.
By Yongliang Wu, Haori Lu, Yulun Wu, Jinqi Luo, Xingyu Zhu, Yaoyao Liu