Research map / Label-efficient and robust learning / Domain adaptation and generalization
One-Shot Unsupervised Cross Domain Translation
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 |
|---|---|---|---|
| Add selective backpropagation, augmentation, and one-way cycle loss instead of using none of them (ablation) | OST · MNIST to SVHN | accuracy: 0.07 -> 0.23 | Helped |
| Add selective backpropagation, augmentation, and one-way cycle loss instead of using none of them (ablation) | OST · SVHN to MNIST | accuracy: 0.10 -> 0.23 | Helped |
| Add selective backpropagation while retaining augmentation and one-way cycle loss (ablation) | OST · MNIST to SVHN | accuracy: 0.14 -> 0.23 | Helped |
| Add selective backpropagation while retaining augmentation and one-way cycle loss (ablation) | OST · SVHN to MNIST | accuracy: 0.14 -> 0.23 | Helped |
| Add augmentation while retaining one-way cycle loss and selective backpropagation (ablation) | OST · MNIST to SVHN | accuracy: 0.22 -> 0.23 | Helped |
| Add augmentation while retaining one-way cycle loss and selective backpropagation (ablation) | OST · SVHN to MNIST | accuracy: 0.23 -> 0.23 | No clear effect |
| Add one-way cycle loss while retaining augmentation and selective backpropagation (ablation) | OST · MNIST to SVHN | accuracy: 0.20 -> 0.23 | Helped |
| Add one-way cycle loss while retaining augmentation and selective backpropagation (ablation) | OST · SVHN to MNIST | accuracy: 0.20 -> 0.23 | Helped |
| Allow phase-II updates to the shared encoder and decoder instead of freezing them (ablation) | OST · MNIST to SVHN | accuracy: 0.16 -> 0.23 | Helped |
| Allow phase-II updates to the shared encoder and decoder instead of freezing them (ablation) | OST · SVHN to MNIST | accuracy: 0.15 -> 0.23 | Helped |
| Use one-way rather than two-way cycle loss (ablation) | OST with selective backpropagation · MNIST to SVHN | accuracy: 0.20 -> 0.23 | Helped |
| Use one-way rather than two-way cycle loss (ablation) | OST with selective backpropagation · SVHN to MNIST | accuracy: 0.13 -> 0.23 | Helped |
Closest papers by meaning
- Unsupervised Cross-Domain Image Generation (ICLR 2017 · similarity 0.88)
- Image-to-image translation for cross-domain disentanglement (NeurIPS 2018 · similarity 0.88)
- Unsupervised Image-to-Image Translation Using Domain-Specific Variational Information Bound (NeurIPS 2018 · similarity 0.88)
- Cycle-consistent Masked AutoEncoder for Unsupervised Domain Generalization (ICLR 2023 · similarity 0.87)
- Towards Identifiable Unsupervised Domain Translation: A Diversified Distribution Matching Approach (ICLR 2024 · similarity 0.87)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 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.