Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning
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 |
|---|---|---|---|
| Replace direct word-embedding or co-occurrence association estimates with learned class-attribute relations | Unsupervised zero-shot classification · AwA | mean per-class accuracy: 67.5 vs 48.6 | Helped |
| Replace direct word-embedding or co-occurrence association estimates with learned class-attribute relations | Unsupervised zero-shot classification · aPaY | mean per-class accuracy: 37.0 vs 37.1 | No clear effect |
| Use data-driven relations instead of semantic relations | Class-attribute association prediction · AwA | mAP: 79.1 -> 79.7 | No clear effect |
| Use data-driven relations instead of semantic relations | Class-attribute association prediction · aPaY | mAP: 76.1 -> 75.7 | No clear effect |
| Transfer and jointly use attributes from both AwA and aPaY | Unsupervised zero-shot classification · aPaY | accuracy: 37.0 -> 49.0 | Helped |
| Transfer and jointly use attributes from both AwA and aPaY | Unsupervised zero-shot classification · AwA | accuracy: 67.5 -> 68.6 | Helped |
| Add a learned hierarchical has_ancestor relation | Class-attribute relation learning · AwA has_pattern relation | mAP change: +2.5% | Helped |
| Add a learned hierarchical has_ancestor relation | Class-attribute relation learning · AwA feeding_type relation | mAP change: +2.1% | Helped |
| Use multiple attribute relations rather than one abstract relation (ablation) | Unsupervised zero-shot classification · AwA | mean per-class accuracy: 5% | Helped |
| Use multiple attribute relations rather than one abstract relation (ablation) | Unsupervised zero-shot classification · aPaY | mean per-class accuracy: 12% | Helped |
| Optimize attribute embeddings during relation learning (ablation) | Unsupervised zero-shot classification · AwA | mean per-class accuracy: 2% | Helped |
| Optimize attribute embeddings during relation learning (ablation) | Unsupervised zero-shot classification · aPaY | mean per-class accuracy: 9% | Helped |
Closest papers by meaning
- Attribute Prototype Network for Zero-Shot Learning (NeurIPS 2020 · similarity 0.90)
- Preserving Semantic Relations for Zero-Shot Learning (CVPR 2018 · similarity 0.89)
- Make an Omelette with Breaking Eggs: Zero-Shot Learning for Novel Attribute Synthesis (NeurIPS 2022 · similarity 0.89)
- Fine-Grained Generalized Zero-Shot Learning via Dense Attribute-Based Attention (CVPR 2020 · similarity 0.89)
- Class Semantic Attribute Perception Guided Zero-Shot Learning (AAAI 2025 · similarity 0.89)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 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.