Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Adversarial Zero-shot Learning With Semantic Augmentation
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 |
|---|---|---|---|
| map into common space instead of intermediate representation space (GANZrl vs GANZrl-IR) (ablation) | GANZrl-IR · CUB | hit@1 accuracy: 55.32 -> 62.56 | Helped |
| add semantically compound samples to GANZrl-SS (ablation) | GANZrl-SS · CUB | hit@1 accuracy: 58.89 -> 62.56 | Helped |
| add semantically same samples to GANZrl-SC (ablation) | GANZrl-SC · AwA | hit@1 accuracy: 82.76 -> 86.23 | Helped |
| use both sample types instead of only compound samples (ablation) | GANZrl-SC · ImageNet-1 | hit@5 accuracy: 61.10 -> 54.95 | Hurt |
| use both sample types instead of only compound samples (ablation) | GANZrl-SC · ImageNet-2 | hit@5 accuracy: 30.80 -> 29.58 | Hurt |
| map into common space instead of intermediate representation space (ablation) | GANZrl-IR · SUN | hit@1 accuracy: 90.31 -> 93.59 | Helped |
| GANZrl vs best prior SAE on small-scale recognition | SAE · AwA | hit@1 accuracy: 84.7 -> 86.23 | Helped |
| GANZrl vs JLSE on image retrieval | JLSE · CUB | MAP: 29.15 -> 60.77 | Helped |
| vary number of unmatched labels K from 5 to 30 (ablation) | GANZrl · small and large-scale datasets | accuracy | No clear effect |
| vary interpolation ratio alpha from 0.1 to 0.9 (ablation) | GANZrl · small and large-scale datasets | accuracy | No clear effect |
Closest papers by meaning
- Generative Dual Adversarial Network for Generalized Zero-Shot Learning (CVPR 2019 · similarity 0.91)
- Synthesized Classifiers for Zero-Shot Learning (CVPR 2016 · similarity 0.90)
- Feature Generating Networks for Zero-Shot Learning (CVPR 2018 · similarity 0.90)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.87)
- On the Zero-shot Adversarial Robustness of Vision-Language Models: A Truly Zero-shot and Training-free Approach (CVPR 2025 · similarity 0.87)
- Class Semantic Attribute Perception Guided Zero-Shot Learning (AAAI 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.