Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Siamese Contrastive Embedding Network for Compositional 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 |
|---|---|---|---|
| Use separate state- and object-specific contrastive spaces instead of the base classifiers (ablation) | SCEN · MIT-States | test AUC: 1.6 -> 4.3 | Helped |
| Add the State Transition Module to the base model (ablation) | SCEN · MIT-States | test AUC: 1.6 -> 3.9 | Helped |
| Combine the contrastive-space loss and State Transition Module (ablation) | SCEN · MIT-States | test AUC: 1.6 -> 5.3 | Helped |
| Use separate state- and object-specific contrastive spaces instead of the base classifiers (ablation) | SCEN · UT-Zappos | test HM: 40.8 -> 44.2 | Helped |
| Add the State Transition Module to the base model (ablation) | SCEN · UT-Zappos | test HM: 40.8 -> 43.9 | Helped |
| Combine the contrastive-space loss and State Transition Module (ablation) | SCEN · UT-Zappos | test HM: 40.8 -> 47.8 | Helped |
| Use SCEN rather than CGE | CZSL · MIT-States test | AUC: 5.1 -> 5.3 | Helped |
| Use SCEN rather than TMN | CZSL · UT-Zappos test | HM: 45.0 -> 47.8 | Helped |
| Use SCEN rather than CGE | CZSL · C-GQA test | AUC: 2.5 -> 2.9 | Helped |
Closest papers by meaning
- Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.84)
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.84)
- Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.83)
- TPR: Topology-Preserving Reservoirs for Generalized Zero-Shot Learning (NeurIPS 2024 · similarity 0.83)
- On the Similarities of Embeddings in Contrastive Learning (ICML 2025 · similarity 0.83)
- A Dynamic Learning Method towards Realistic Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.82)
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.