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
arXiv:2605. 19350v2 Announce Type: replace-cross Abstract: Creating and editing high-quality 3D content remains a central challenge in computer graphics.
By Habib Slim, Shariq Farooq Bhat, Mohamed Elhoseiny, Yifan Wang, Mike Roberts
arXiv:2607. 16409v1 Announce Type: cross Abstract: Unified Multimodal Large Language Models (MLLMs) offer a promising paradigm for unifying visual understanding and generation, yet they still struggle to follow complex spatial instructions and logical constraints in controllable image generation.
By Junhao Liu, Jian-Wei Zhang, Tao Huang, Miles Yang, Zhao Zhong, Liefeng Bo
Image outpainting extends an image beyond its original borders, requiring seamless style integration and globally coherent scene completion. Building on the success of diffusion models, recent methods have achieved substantial improvements in visual quality.
MOSS-Image is an open-weight image generation system that combines a 5B single-stream diffusion transformer with an intention-aware prompt enhancement module to produce high-fidelity, text-rich images with precise layout control. The system supports bilingual long-form text rendering and introduces evaluation benchmarks LongLongTextBench and RealLayoutArena. It achieves state‑of‑the‑art performance on LLTB and ranks highly on RLA, with plans to release the model, training recipe, and evaluation framework for the community.
By MOSI. AI
arXiv:2608.20448v1 Announce Type: cross
Abstract: Digital 3D objects used in games and animation are often required to be compositional; that is, decomposed into semantically meaningful parts. Recent...
By Ava Pun, Kangle Deng, Yiheng Zhu, Jun-Yan Zhu, Maneesh Agrawala, Tinghui Zhou
arXiv:2602.17690v3 Announce Type: replace-cross
Abstract: Graphic design generation demands a delicate balance between high visual fidelity and fine-grained structural editability. However, existing...
By Ziyuan Liu, Shizhao Sun, Danqing Huang, Yingdong Shi, Meisheng Zhang, Ji Li, Jingsong Yu, Jiang Bian
arXiv:2502. 06819v2 Announce Type: replace Abstract: This paper presents a framework for generating 3D indoor scenes from text prompts.
By Yao Wei, Matteo Toso, Pietro Morerio, Changjae Oh, Michael Ying Yang, Alessio Del Bue
Sphinx is a training‑free hybrid inference system for novel view synthesis that combines fast regression models with selective diffusion refinement. It first generates coarse views via regression, then applies diffusion only where needed to improve quality while reducing computation. An SLO‑aware scheduler dynamically allocates GPU resources, achieving up to 2.7× faster inference with less than 10% perceptual loss and up to 10× higher SLO attainment compared to FCFS scheduling.
By Yuchen Xia, Souvik Kundu, Mosharaf Chowdhury, Nishil Talati
arXiv:2603. 28762v2 Announce Type: replace-cross Abstract: Modern Text-to-Image (T2I) diffusion models have achieved remarkable semantic alignment, yet they often suffer from a significant lack of variety, converging on a narrow set of visual solutions for any given prompt.
By Omer Dahary, Benaya Koren, Daniel Garibi, Daniel Cohen-Or
The paper introduces a new sequential text‑to‑scene generation framework that jointly produces scene layouts and objects. It proposes a 3D Autoregressive Diffusion model (3D‑ARD+) that first generates coarse 3D latents conditioned on text and the current scene, then refines them into fine‑grained object geometry and appearance. A large dataset of 230K indoor scenes with paired text instructions is curated for training, and the model demonstrates the ability to place objects according to complex spatial and semantic text prompts.
By Zhenggang Tang, Yuehao Wang, Yuchen Fan, Jun-Kun Chen, Yu-Ying Yeh, Kihyuk Sohn, Zhangyang Wang, Qixing Huang, Alexander Schwing, Rakesh Ranjan, Dilin Wang, Zhicheng Yan
arXiv:2601.21857v2 Announce Type: replace
Abstract: We revisit diffusion-based generation for structured visual content and identify a fundamental limitation of existing approaches: foreground preser...
By Taewon Kang, Yu Shen, Ming C. Lin