Research map / Image generation / GANs
Variational Diffusion Autoencoders with Random Walk Sampling
Henry Li, Ofir Lindenbaum, Xiuyuan Cheng, Alexander Cloninger
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use VDAE instead of WGAN | generative model · Bulldog | Gromov-Wasserstein distance: 117.3 -> 15.3 | Improved |
| use VDAE instead of VAE | generative model · Bulldog | FID: 245.7 -> 144.3 | Improved |
| use VDAE instead of GAN | generative model · Bulldog | MMD: 741.8 -> 9.68 | Improved |
| use VDAE instead of GAN | generative model · Stanford bunny | MMD: 175.6 -> 3.6 | Improved |
| use VDAE instead of WGAN | generative model · MNIST | Gromov-Wasserstein distance: 3.6 -> 14.4 | Hurt |
| use VDAE instead of beta-VAE | generative model · MNIST | Median bi-Lipschitz: 1.6 -> 8.4 | Hurt |
| use VDAE instead of SVAE | generative model · Frey faces | MMD: 29.4 -> 47.0 | Hurt |
| use VDAE instead of WGAN | generative model · Torus | Median bi-Lipschitz: 3.3 -> 4.8 | Hurt |
| use VDAE instead of beta-VAE | generative model · Stanford bunny | Median bi-Lipschitz: 34.4 -> 5.5 | Improved |
| repeat random-walk resampling steps | VDAE sampling · synthetic manifolds | convergence to uniform distribution: 20 steps; stationary within 5-10 iterations | Improved |
Closest papers by meaning
- What Matters for Diffusion-Friendly Latent Manifold? Prior-Aligned Autoencoders for Latent Diffusion (NeurIPS 2026 · similarity 0.89)
- Generalization in VAE and Diffusion Models: A Unified Information-Theoretic Analysis (ICLR 2025 · similarity 0.88)
- Structure-Guided Adversarial Training of Diffusion Models (CVPR 2024 · similarity 0.88)
- Unified Generation, Reconstruction, and Representation: Generalized Diffusion with Adaptive Latent Encoding-Decoding (ICML 2024 · similarity 0.88)
- DiffEnc: Variational Diffusion with a Learned Encoder (ICLR 2024 · similarity 0.88)
- Revisiting Diffusion Models: From Generative Pre-training to One-Step Generation (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
