Iterative Flow Matching: Path Correction and Gradual Refinement for Enhanced Generative Modeling
Abstract
Abstract. Generative models for image generation are now commonly used for a wide variety of applications, ranging from guided image generation for entertainment to solving inverse problems. Nonetheless, training a generator is a nontrivial feat that requires fine-tuning and can lead to so-called hallucinations, that is, the generation of images that are unrealistic. In this work, we explore image generation using flow matching. We explain and demonstrate why flow matching can generate hallucinations, and we propose an iterative process to improve the generation process. Our iterative process can be integrated into virtually any generative modeling technique, thereby enhancing the performance and robustness of image synthesis systems.
Fields of Study & Subjects
Generative Adversarial Networks and Image Synthesis
Image Enhancement Techniques
Music Technology and Sound Studies
Publication details
- Journal / source
- SIAM Journal on Scientific Computing
- Volume
- 48
- Issue
- 4
- Pages
- C814–C831
- Publication date
- 20 July 2026
- Language
- EN