Research map / Probabilistic and scientific modeling / Variational and generative sampling
Neural Discrete Representation Learning
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 |
|---|---|---|---|
| Use discrete vector-quantised latents instead of continuous Gaussian latents | VQ-VAE · CIFAR10 | bits/dim: 4.51 -> 4.67 | Hurt |
| Use VQ-VAE instead of VIMCO | CIFAR10 · CIFAR10 | bits/dim: 5.14 -> 4.67 | Helped |
| Vary the commitment-loss weight β from 0.1 to 2.0 (ablation) | VQ-VAE · CIFAR10 | bits/dim | No clear effect |
| Use a discrete latent code instead of a random latent space | VQ-VAE · VCTK phoneme mapping | 41-way classification accuracy: 7.2 % -> 49.3 % | Helped |
| Condition video generation on actions (ablation) | VQ-VAE · DeepMind Lab video | visual quality | No clear effect |
| Use discrete latents with a powerful PixelCNN decoder | VQ-VAE · DeepMind Lab frames | latent usage | Helped |
Closest papers by meaning
- Vector Quantized Wasserstein Auto-Encoder (ICML 2023 · similarity 0.92)
- Diffusion bridges vector quantized variational autoencoders (ICML 2022 · similarity 0.91)
- Finite Scalar Quantization: VQ-VAE Made Simple (ICLR 2024 · similarity 0.88)
- Restructuring Vector Quantization with the Rotation Trick (ICLR 2025 · similarity 0.88)
- Learning to Quantize for Training Vector-Quantized Networks (ICML 2025 · similarity 0.88)
- Information-theoretic Generalization Analysis for VQ-VAEs: A Role of Latent Variables (NeurIPS 2025 · similarity 0.87)
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.