xie jiake — accepted AI-conference papers

2025

IterComp: Iterative Composition-Aware Feedback Learning from Model Gallery for Text-to-Image Generation

Xinchen Zhang, Ling Yang, Guohao Li, YaQi Cai, xie jiake, Yong Tang +3

ICLR 2025poster

Advanced diffusion models like Stable Diffusion 3, Omost, and FLUX have made notable strides in compositional text-to-image generation. However, these methods typically exhibit distinct strengths for compositional generation, with some excelling in handling attribute binding and others in spatial re…

Cited by 12 Source ↗PDF ↗code ↗Scholar ↗

2024

RealCompo: Balancing Realism and Compositionality Improves Text-to-Image Diffusion Models

Xinchen Zhang, Ling Yang, YaQi Cai, Zhaochen Yu, Kai-Ni Wang, xie jiake +5

NeurIPS 2024poster

Diffusion models have achieved remarkable advancements in text-to-image generation. However, existing models still have many difficulties when faced with multiple-object compositional generation. In this paper, we propose ***RealCompo***, a new *training-free* and *transferred-friendly* text-to-imag…

Cited by 7 Source ↗PDF ↗code ↗Scholar ↗