Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Conditional Attributes for Compositional 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 |
|---|---|---|---|
| use conditional attribute embeddings based on recognized object and image | CANet · UT-Zappos50K | AUC: 32.0 -> 33.1 | Helped |
| use conditional attribute embeddings based on recognized object and image | CANet · MIT-States | best unseen: 25.2 -> 26.2 | Helped |
| use conditional attribute embeddings based on recognized object and image | CANet · C-GQA | best attr: 13.6 -> 17.5 | Helped |
| remove attribute and object classification losses (ablation) | CANet · UT-Zappos50K | AUC: 33.1 -> 31.7 | Hurt |
| remove attribute and object classification losses (ablation) | CANet · C-GQA | AUC: 3.3 -> 3.0 | Hurt |
| remove composition classification loss (ablation) | CANet · UT-Zappos50K | AUC: 33.1 -> 28.3 | Hurt |
| remove composition classification loss (ablation) | CANet · C-GQA | AUC: 3.3 -> 2.6 | Hurt |
| remove the conditional attribute framework (ablation) | CANet · UT-Zappos50K | AUC: 33.1 -> 31.7 | Hurt |
| remove the conditional attribute framework (ablation) | CANet · C-GQA | AUC: 3.3 -> 3.1 | Hurt |
| remove the prior composing network P (ablation) | CANet · UT-Zappos50K | AUC: 33.1 -> 31.6 | Hurt |
| remove the prior composing network P (ablation) | CANet · C-GQA | AUC: 3.3 -> 3.1 | Hurt |
| remove the attribute hyper learner H (ablation) | CANet · UT-Zappos50K | AUC: 33.1 -> 30.5 | Hurt |
Closest papers by meaning
- Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning (CVPR 2024 · similarity 0.93)
- Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.92)
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.92)
- A Dynamic Learning Method towards Realistic Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.91)
- LOGICZSL: Exploring Logic-induced Representation for Compositional Zero-shot Learning (CVPR 2025 · similarity 0.89)
- Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (AAAI 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.