fid
Fréchet Inception Distance, a metric for assessing the quality of generated images by comparing the distribution of real and generated images in the feature space of a pretrained model.
- Balanced Conic Rectified Flow
- Diffusion Tree Sampling: Scalable inference‑time alignment of diffusion models
- Latent Zoning Network: A Unified Principle for Generative Modeling, Representation Learning, and Classification
- Learning to Integrate Diffusion ODEs by Averaging the Derivatives
- Mean Flows for One-step Generative Modeling
- Mean Flows for One-step Generative Modeling
- NFIG: Multi-Scale Autoregressive Image Generation via Frequency Ordering
- TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics
- Token Perturbation Guidance for Diffusion Models
- When Worse is Better: Navigating the Compression Generation Trade-off In Visual Tokenization