Research map / Label-efficient and robust learning / Domain adaptation and generalization
Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation
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 CDRD instead of ADGAN | UDA digit classification · MNIST→USPS | accuracy (%): 92.50 -> 95.05 | Helped |
| Use CDRD instead of ADGAN | UDA digit classification · USPS→MNIST | accuracy (%): 90.80 -> 94.35 | Helped |
| Use CDRD instead of UNIT | UDA face classification · CelebA sketch→photo (smiling) | accuracy (%): 81.04 -> 87.61 | Helped |
| Add the encoder and VAE translation objective to CDRD | CelebA sketch→photo (smiling) · UDA face classification | accuracy (%): 87.61 -> 88.28 | Helped |
| Use CDRD instead of UNIT | UDA face classification · CelebA sketch→photo (glasses) | accuracy (%): 79.89 -> 94.49 | Helped |
| Add the encoder and VAE translation objective to CDRD | CelebA sketch→photo (glasses) · UDA face classification | accuracy (%): 94.49 -> 94.84 | Helped |
| Use CDRD instead of UNIT | UDA scene classification · Scene photo→paint (night) | accuracy (%): 67.81 -> 84.21 | Helped |
| Add the encoder and VAE translation objective to CDRD | Scene photo→paint (night) · UDA scene classification | accuracy (%): 84.21 -> 85.58 | Helped |
| Use CDRD instead of UNIT | UDA scene classification · Scene photo→paint (season) | accuracy (%): 66.09 -> 79.87 | Helped |
| Add the encoder and VAE translation objective to CDRD | Scene photo→paint (season) · UDA scene classification | accuracy (%): 79.87 -> 80.03 | Helped |
| Set disentanglement weight λ too low (ablation) | CDRD and E-CDRD image synthesis · Conditional attribute manipulation | attribute manipulability | Hurt |
| Set disentanglement weight λ extremely high (ablation) | CDRD and E-CDRD image synthesis · Conditional image quality | image quality | Hurt |
Closest papers by meaning
- Diffusion Model with Cross Attention as an Inductive Bias for Disentanglement (NeurIPS 2024 · similarity 0.88)
- Disentangled Representation Learning via Flow Matching (NeurIPS 2026 · similarity 0.87)
- Disentangled Representation Learning via Modular Compositional Bias (NeurIPS 2025 · similarity 0.87)
- Transferring disentangled representations: bridging the gap between synthetic and real images (NeurIPS 2024 · similarity 0.87)
- Triple Feature Disentanglement for One-Stage Adaptive Object Detection (AAAI 2024 · similarity 0.86)
- An Information Criterion for Controlled Disentanglement of Multimodal Data (ICLR 2025 · similarity 0.86)
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.