Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Self-Supervised Domain-Aware Generative Network for Generalized Zero-Shot Learning
Check 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 SDGN instead of GXE | GZSL · CUB | H: 62.3 -> 70.1 | Helped |
| Use SDGN instead of GXE | GZSL · AwA1 | H: 88.4 -> 87.7 | Hurt |
| Use SDGN instead of f-VAEGAN | GZSL · AwA2 | H: 86.7 -> 89.1 | Helped |
| Use SDGN instead of GXE | GZSL · SUN | H: 51.0 -> 52.8 | Helped |
| Use SDGN instead of f-VAEGAN | GZSL · FLO | H: 82.7 -> 84.4 | Helped |
| Add target-domain discriminator Dt (ablation) | G + Ds on CUB · CUB | H: 51.0 -> 68.5 | Helped |
| Add target-domain discriminator Dt (ablation) | G + Ds on AwA2 · AwA2 | H: 60.0 -> 86.9 | Helped |
| Add target-domain discriminator Dt (ablation) | G + Ds on SUN · SUN | H: 39.4 -> 50.2 | Helped |
| Add SLM and cross-domain triplet mining (ablation) | G + Ds + Dt on CUB · CUB | H: 68.5 -> 70.1 | Helped |
| Add SLM and cross-domain triplet mining (ablation) | G + Ds + Dt on AwA2 · AwA2 | H: 86.9 -> 89.1 | Helped |
| Add SLM and cross-domain triplet mining (ablation) | G + Ds + Dt on SUN · SUN | H: 50.2 -> 52.8 | Helped |
| Increase the SLM loss coefficient λa above 0.1 (ablation) | SDGN on CUB · CUB | accuracy | Hurt |
Closest papers by meaning
- Adaptive and Generative Zero-Shot Learning (ICLR 2021 · similarity 0.90)
- Gradient Matching Generative Networks for Zero-Shot Learning (CVPR 2019 · similarity 0.90)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.90)
- Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition (AAAI 2024 · similarity 0.88)
- GenZSL: Generative Zero-Shot Learning Via Inductive Variational Autoencoder (ICML 2025 · similarity 0.88)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 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.