| pipeline_tag: text-to-image | |
| library_name: transformers | |
| license: mit | |
| This model was fine-tuned on Qwen to generate reflections in the test-time scaling framework introduced in [From Reflection to Perfection: Scaling Inference-Time Optimization for Text-to-Image Diffusion Models via Reflection Tuning](https://huggingface.co/papers/2504.16080). This model enhances text-to-image generation by incorporating reflection tuning. | |
| For more details and usage instructions, please refer to the paper and the [code repository](https://github.com/Diffusion-CoT/ReflectionFlow). The project page can be found [here](https://diffusion-cot.github.io/reflection2perfection). |