State of the Field
Recent advancements in generative image modeling are focusing on enhancing the precision and efficiency of text-to-image generation. Techniques like CoCo leverage executable code for structured scene planning, improving the fidelity of generated images, while InnoAds-Composer streamlines e-commerce poster creation through tri-conditional control, addressing issues of subject fidelity and style consistency. Meanwhile, frameworks such as Hierarchical Concept-to-Appearance Guidance and Layer-wise Instance Binding are tackling challenges in multi-subject generation and regional control, ensuring better identity preservation and occlusion management. Innovations like Reflective Flow Sampling and ArcFlow are reducing inference costs by optimizing the distillation process, allowing for faster generation without compromising quality. The field is increasingly prioritizing user control and adaptability, with methods designed to support editable workflows and real-time modifications, making generative models more applicable to commercial needs in advertising, content creation, and interactive applications.
Papers
1–10 of 10Hierarchical Concept-to-Appearance Guidance for Multi-Subject Image Generation
Multi-subject image generation aims to synthesize images that faithfully preserve the identities of multiple reference subjects while following textual instructions. However, existing methods often su...
InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generation
E-commerce product poster generation aims to automatically synthesize a single image that effectively conveys product information by presenting a subject, text, and a designed style. Recent diffusion ...
CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generation
Recent advancements in Unified Multimodal Models (UMMs) have significantly advanced text-to-image (T2I) generation, particularly through the integration of Chain-of-Thought (CoT) reasoning. However, e...
The Latent Color Subspace: Emergent Order in High-Dimensional Chaos
Text-to-image generation models have advanced rapidly, yet achieving fine-grained control over generated images remains difficult, largely due to limited understanding of how semantic information is e...
HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion Acceleration
Diffusion models have demonstrated a remarkable ability in Text-to-Image (T2I) generation applications. Despite the advanced generation output, they suffer from heavy computation overhead, especially ...
Layer-wise Instance Binding for Regional and Occlusion Control in Text-to-Image Diffusion Transformers
Region-instructed layout control in text-to-image generation is highly practical, yet existing methods suffer from limitations: (i) training-based approaches inherit data bias and often degrade image ...
TIDE: Text-Informed Dynamic Extrapolation with Step-Aware Temperature Control for Diffusion Transformers
Diffusion Transformer (DiT) faces challenges when generating images with higher resolution compared at training resolution, causing especially structural degradation due to attention dilution. Previou...
ArcFlow: Unleashing 2-Step Text-to-Image Generation via High-Precision Non-Linear Flow Distillation
Diffusion models have achieved remarkable generation quality, but they suffer from significant inference cost due to their reliance on multiple sequential denoising steps, motivating recent efforts to...
WaDi: Weight Direction-aware Distillation for One-step Image Synthesis
Despite the impressive performance of diffusion models such as Stable Diffusion (SD) in image generation, their slow inference limits practical deployment. Recent works accelerate inference by distill...
Reflective Flow Sampling Enhancement
The growing demand for text-to-image generation has led to rapid advances in generative modeling. Recently, text-to-image diffusion models trained with flow matching algorithms, such as FLUX, have ach...