Research map / Probabilistic and scientific modeling / Variational and generative sampling
Glow: Generative Flow with Invertible 1x1 Convolutions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace channel reversal with a learned invertible 1x1 convolution (ablation) | Glow flow steps with additive coupling · CIFAR-10 | negative log-likelihood | Helped |
| Replace fixed random channel permutation with a learned invertible 1x1 convolution (ablation) | Glow flow steps with additive coupling · CIFAR-10 | negative log-likelihood | Helped |
| Replace channel reversal with a learned invertible 1x1 convolution (ablation) | Glow flow steps with affine coupling · CIFAR-10 | negative log-likelihood | Helped |
| Replace fixed random channel permutation with a learned invertible 1x1 convolution (ablation) | Glow flow steps with affine coupling · CIFAR-10 | negative log-likelihood | Helped |
| Use affine coupling instead of additive coupling (ablation) | Glow · CIFAR-10 | convergence speed | Helped |
| Increase model depth from L = 4 to L = 6 (ablation) | Glow · CelebA-HQ | sample quality and long-range dependencies | Helped |
| Lower sampling temperature from 1.0 toward 0.7 (ablation) | Glow · CelebA-HQ | sample quality and diversity | Mixed |
| Replace RealNVP with Glow | Generative image modeling · CIFAR-10, ImageNet, and LSUN | bits per dimension: CIFAR-10: 3.49 -> 3.35; ImageNet 32x32: 4.28 -> 4.09; ImageNet 64x64: 3.98 -> 3.81; LSUN bedroom: 2.72 -> 2.38; LSUN tow | Helped |
Closest papers by meaning
- Mean Flows for One-step Generative Modeling (NeurIPS 2025 · similarity 0.86)
- Contextual Flow Matching: Adaptive Step Selection in Flow Models for Efficient Visual Generation (NeurIPS 2026 · similarity 0.85)
- Self-Corrected Flow Distillation for Consistent One-Step and Few-Step Image Generation (AAAI 2025 · similarity 0.85)
- Scaling Rectified Flow Transformers for High-Resolution Image Synthesis (ICML 2024 · similarity 0.85)
- Text-to-Image Rectified Flow as Plug-and-Play Priors (ICLR 2025 · similarity 0.85)
- STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis (NeurIPS 2025 · similarity 0.85)
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.