Research map / Vision-language and multimodal / CLIP and zero-shot
Prototype-Conditioned Imagination for Compositional Zero-Shot Learning
Yifan Zhu, Haofeng Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 Visual Enrichment (VE) to 3-branch baseline (ablation) | PCI BAD · UT-Zappos (closed-world) | AUC: 42.1 -> 44.2 | Improved |
| add Semantic Disentanglement (SD) to 3-branch baseline (ablation) | PCI BAD · UT-Zappos (closed-world) | AUC: 42.1 -> 43.2 | Improved |
| combine VE and SD (BAD) (ablation) | PCI BAD · UT-Zappos (closed-world) | AUC: 42.1 -> 47.2 | Improved |
| add VE and SD (ablation) | PCI BAD · MIT-States (closed-world) | AUC: 22.3 -> 23.4 | Improved |
| add PCM on top of BAD (ablation) | PCI · UT-Zappos (closed-world) | AUC: 47.2 -> 49.0 | Improved |
| add PCM on top of BAD (ablation) | PCI · MIT-States (closed-world) | AUC: 23.4 -> 24.1 | Improved |
| add PCM on top of BAD (ablation) | PCI · UT-Zappos (open-world) | AUC: 40.0 -> 42.1 | Improved |
| add PCM on top of BAD (ablation) | PCI · MIT-States (open-world) | AUC: 9.1 -> 9.6 | Improved |
| use OT-based SD instead of vanilla multi-head attention (ablation) | PCI BAD · UT-Zappos (closed-world) | AUC: 45.8 -> 47.2 | Improved |
| add OT distributional regularizer Lot to SD (ablation) | PCI BAD · UT-Zappos (closed-world) | AUC: 46.7 -> 47.2 | Improved |
| add OT distributional regularizer Lot to SD (ablation) | PCI BAD · MIT-States (open-world) | AUC: 8.8 -> 9.1 | Improved |
| use compositional modulation instead of Add fusion (ablation) | PCI PCM · UT-Zappos (closed-world) | AUC: 47.4 -> 49.0 | Improved |
Closest papers by meaning
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.90)
- ProCC: Progressive Cross-Primitive Compatibility for Open-World Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.89)
- Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning (CVPR 2024 · similarity 0.88)
- Unlocking Compositional Generalization in Continual Few-Shot Learning (NeurIPS 2026 · similarity 0.88)
- A Dynamic Learning Method towards Realistic Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.88)
- Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (AAAI 2024 · 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.
