Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Semantic Autoencoder for 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 |
|---|---|---|---|
| add the visual-feature reconstruction constraint to visual-to-semantic ridge regression (ablation) | SAE · AwA, CUB, aP&Y, SUN | accuracy (%): 60.6 -> 84.7; 41.1 -> 61.4; 30.5 -> 55.4; 71.5 -> 91.0 | Helped |
| add the visual-feature reconstruction constraint to semantic-to-visual ridge regression (ablation) | SAE · AwA, CUB, aP&Y, SUN | accuracy (%): 80.4 -> 84.7; 52.4 -> 61.4; 48.8 -> 55.4; 84.5 -> 91.0 | Helped |
| use the encoder projection W rather than decoder projection Wᵀ | SAE · small-scale ZSL benchmarks | accuracy (%): AwA 84.7 vs 84.0; CUB 61.4 vs 60.9; aP&Y 55.4 vs 54.8; SUN 91.0|65.2 vs 91.5|65.2 | Mixed |
| use SAE rather than SynC struct | generalized ZSL · AwA | AUSUC: 0.583 -> 0.579 | Hurt |
| use SAE rather than SynC struct | generalized ZSL · CUB | AUSUC: 0.356 -> 0.448 | Helped |
| use SAE rather than ESZSL | zero-shot learning on AwA · AwA | training time (seconds): 16 -> 1.3 | Helped |
| add the semantic decoder/reconstruction constraint to supervised clustering | SAE · Oxford Flowers-17 segmentation | test loss: 1.29±0.01 -> 1.19±0.01 | Helped |
Closest papers by meaning
- GenZSL: Generative Zero-Shot Learning Via Inductive Variational Autoencoder (ICML 2025 · similarity 0.91)
- Learning a Deep Embedding Model for Zero-Shot Learning (CVPR 2017 · similarity 0.91)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.90)
- Data-Free Generalized Zero-Shot Learning (AAAI 2024 · similarity 0.89)
- ZeroMamba: Exploring Visual State Space Model for Zero-Shot Learning (AAAI 2025 · similarity 0.89)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.89)
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.