1. LATENT CONSISTENCY MODELS: SYNTHESIZING HIGH-RESOLUTION IMAGES WITH FEW-STEP INFERENCE
2. Abstract
3. 1 Introduction
4. 2 Related Work
5. 3 Preliminaries
6. 4 Latent Consistency Models
- 4.1 Consistency Distillation in the Latent Space
- 4.2 One-Stage Guided Distillation by solving augmented PF-ODE
- 4.3 Accelerating Distillation with Skipping Time Steps
- 4.4 Latent Consistency Fine-tuning for customized dataset
7. 5 Experiment
- 5.1 Text-to-Image Generation
- 5.2 Ablation Study
- 5.3 Downstream Consistency Fine-tuning Results
8. 6 Conclusion
9. More Details on Diffusion and Consistency Models
- A Diffusion Models
- B More Details on Consistency Models in song2023consistency
10. Multistep Latent Consistency Sampling
11. Algorithm Details of Latent Consistency Fine-tuning
12. Different ways to parameterize the consistency function
13. Formulas of Other ODE Solvers
14. Training Details of Latent Consistency Distillation
15. Reproduction Details of Guided-Distill
16. More few-step Inference Results
17. References
18. Appendix
- A More Details on Diffusion and Consistency Models
- B Multistep Latent Consistency Sampling
- C Algorithm Details of Latent Consistency Fine-tuning
- D Different ways to parameterize the consistency function
- E Formulas of Other ODE Solvers
- F Training Details of Latent Consistency Distillation
- G Reproduction Details of Guided-Distill
- H More Few-Step Inference Results
