Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Invariant Visual Representations for Compositional Zero-Shot Learning
Tian Zhang, Kongming Liang, Ruoyi Du, Xian Sun, Zhanyu Ma, Jun Guo
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 L_cls to L_comp (ablation) | L_comp baseline · UT-Zappos50K | AUC: 27.8 -> 28.8 | Mixed |
| add L_cls to L_comp (ablation) | L_comp baseline · Clothing16K | AUC: 82.9 -> 84.8 | Improved |
| add L_rep to L_cls+L_comp (ablation) | L_cls+L_comp · UT-Zappos50K | AUC: 28.8 -> 29.8 | Improved |
| add L_rep to L_cls+L_comp (ablation) | L_cls+L_comp · Clothing16K | AUC: 84.8 -> 85.3 | Improved |
| add L_grad to L_cls+L_comp (ablation) | L_cls+L_comp · UT-Zappos50K | AUC: 28.8 -> 30.0 | Improved |
| add L_grad to L_cls+L_comp (ablation) | L_cls+L_comp · Clothing16K | AUC: 84.8 -> 86.3 | Improved |
| add L_grad to L_cls+L_comp+L_rep (ablation) | full method · UT-Zappos50K | AUC: 29.8 -> 30.6 | Improved |
| add L_rep to L_cls+L_comp+L_grad (ablation) | full method · Clothing16K | AUC: 86.3 -> 87.0 | Improved |
| add L_rep to L_cls+L_comp (ablation) | L_cls+L_comp · UT-Zappos50K | Seen accuracy: 57.0 -> 55.9 | Hurt |
| use Euclidean instead of cosine gradient distance (ablation) | L_grad · UT-Zappos50K | AUC: 28.6 -> 30.6 | Improved |
| use Euclidean instead of cosine gradient distance (ablation) | L_grad · Clothing16K | AUC: 85.9 -> 87.0 | Improved |
| increase discard ratio alpha from 1/6 to 1/2 (ablation) | channel mask · UT-Zappos50K | AUC: 30.6 -> 28.5 | Hurt |
Closest papers by meaning
- Disentangling Before Composing: Learning Invariant Disentangled Features for Compositional Zero-Shot Learning (TPAMI 2024 · similarity 0.95)
- Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.93)
- A Dynamic Learning Method towards Realistic Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.92)
- Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning (CVPR 2024 · similarity 0.91)
- Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.91)
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.90)
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.
