Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Semantically Aligned Bias Reducing 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 the semantically aligned latent representation and early-stopped generator | SABR-I · AWA2, CUB, and SUN conventional ZSL | accuracy: 65.2 vs 63.8; 63.9 vs 58.8; 62.8 vs 61.4 | Helped |
| Use unseen-data marginal alignment with weak conditional transfer | SABR-T · AWA2, CUB, and SUN conventional ZSL | accuracy: 88.9 vs 79.7; 74.0 vs 72.1; 67.5 vs 63.9 | Helped |
| Use the semantically aligned latent representation and early-stopped generator | SABR-I · AWA2 and CUB GZSL | H: 46.9 vs 32.3; 56.8 vs 49.7 | Helped |
| Use the semantically aligned latent representation and early-stopped generator | SABR-I · SUN GZSL | H: 41.5 vs 39.4 | Helped |
| Use unseen-data marginal alignment with weak conditional transfer | SABR-T · AWA2 and SUN GZSL | H: 85.0 vs 77.4; 48.6 vs 38.8 | Helped |
| Use unseen-data marginal alignment with weak conditional transfer | SABR-T · CUB GZSL | H: 70.3 vs 73.2 | Hurt |
| Stop generator training at the best validation harmonic mean rather than training to convergence (ablation) | SABR-I · AWA2 GZSL validation | H | Helped |
| Use unseen-class unlabeled data to model the marginal distribution (ablation) | SABR-T · AWA2 and SUN transductive ZSL and GZSL | accuracy: 6.5%; 10%; 35% | Helped |
| Tune the weak-transfer weight instead of setting transfer to zero or forcing identical generators (ablation) | SABR-T · AWA2, CUB, and SUN cross-validation | H: 0.008; 0.002; 0.002 | Helped |
Closest papers by meaning
- When Generalized Zero-Shot Learning Meets PU Learning: A Plug-and-Play Framework for Seen-Class Bias Mitigation (ICML 2026 · similarity 0.91)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.91)
- Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition (AAAI 2024 · similarity 0.90)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.89)
- Data-Free Generalized Zero-Shot Learning (AAAI 2024 · similarity 0.89)
- Class Semantic Attribute Perception Guided Zero-Shot Learning (AAAI 2025 · similarity 0.88)
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.