HimmBO is a two‑stage method for preference‑driven merging of large collections of lightweight adapters used to fine‑tune text‑to‑image models. The first stage screens adapters via level‑set estimation on pre‑generated images, avoiding costly image generation, while the second stage optimizes merge coefficients in the reduced subspace using a style‑aware kernel. Experiments show HimmBO attains higher similarity with far fewer image generations than prior methods and scales robustly to collections of up to 1,080 adapters.
arXiv:2610.09015v1 Announce Type: new
Abstract: Diffusion models can generate high-quality images, yet aligning their outputs with individual user preferences remains challenging. A key bottleneck is...
By Jiamu Bai, Jiaming Hu, Yanhong Wu, Zellux Wang
Reference-based diffusion models enable highly controllable image generation by leveraging elements from input images to guide prompt-driven synthesis. However, these models are computationally expensive in runtime, and their cost scales severely with the number of input references.
Modern black-box Image-to-Video (I2V) models offer powerful capabilities in automated content creation, yet their lack of fine-grained control and reliability presents significant challenges in professional workflows. Their inherent stochasticity causes minor variations in textual prompts or hyperparameters to yield drastically different outputs often necessitating inefficient, brute-force trial-and-error processes.
The paper introduces InterIL, a joint generative model that simultaneously produces a background image and a layout of foreground elements for graphic design templates, addressing the limitations of sequential generation approaches. InterIL connects pretrained image and layout diffusion backbones via a learnable communication module, freezing the backbones to preserve prior knowledge while training only the interaction module. The model also offers a test‑time guidance strategy, enabling users to impose preferences without retraining, and demonstrates superior image, layout, and harmonization quality compared to previous methods.
By Shirong Yang, Bo Yang, Ying Cao
EditJev is a System One routing model that selects the best image editing model from a candidate pool for a given image and instruction. It predicts sample-level preference scores in a single forward pass using a lightweight head on a vision‑language backbone, avoiding autoregressive generation or parsing. By combining these scores with model speed metrics, EditJev improves overall editing quality while cutting end‑to‑end latency by up to 60.8%.
By Zhile Guan, Xu Jiang, Yule Duan, Gongbo Zhang, Shuyu Wang, Xiao Chen, Jian Zhang
arXiv:2606. 08841v1 Announce Type: new Abstract: Text-to-image diffusion models are increasingly deployed in open-ended creative contexts, yet their outputs remain impersonal, optimized for aggregate aesthetics rather than individual taste.
By Harini SI, Somesh Singh, Yaman Kumar Singla, David Doermann, Rajiv Ratn Shah
Text-to-image (T2I) platforms typically return several images per prompt so users can explore different visual interpretations and choose one that matches their intent. However, T2I models are usually...
Recent text-to-image models such as DALLE-3 excel at following diverse prompts yet remain blind to individual aesthetic preferences. We study personalized image generation, where models must align outputs with a user's implicit visual preferences based on a few historically preferred images and a short prompt.
How can we personalize a shared expert library from a user's pairwise choices? Prior work can realize different reward trade-offs by merging reward-specialized experts, given a vector of trade-off wei...
T-DIG is a text-conditioned diversity framework that enables users to generate multiple images from a single prompt and then request variations along specific axes—such as background, style, or clothing—using natural language. The system employs an LLM policy to produce K axis-conditioned prompts or edit instructions for a frozen image generator or editor, trained with GRPO and a diversity reward. T‑DIG works with open‑weight, distilled, and proprietary models, and the authors release a new dataset of human‑written axes and real user interactions to benchmark text‑conditioned diversity.
By Simran Khanuja, Rohit Gandikota, Gaurav Parmar, Graham Neubig
Kingfisher is a fast, model‑tuning‑free framework for personalized image generation using visual autoregressive models. It decomposes early autoregressive scale codes into content and style components conditioned on prompts, then recombines these components from two real images into a compact semantic prefix. The frozen generator refines the remaining scales, achieving strong content preservation and superior style and text alignment while cutting pair‑specific preparation time by 54× compared with diffusion methods and 7.5× compared with other VAR approaches.
By Dinh-Khoi Vo, Akihiro Sugimoto, Minh-Triet Tran, Trung-Nghia Le