Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Clustering-based Prototypes for Compositional Zero-Shot Learning
Hongyu Qu, Jianan Wei, Xiangbo Shu, Wenguan Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes ClusPro, which clusters within each attribute/object to find diverse prototypes and uses prototype-based contrastive and HSIC decorrelation losses to shape a disentangled primitive embedding space for compositional zero-shot learning.
Problem
Compositional zero-shot learning must recognize unseen attribute-object pairs, but existing methods represent each primitive with a single centroid prototype and disentangle attributes and objects only from local (batch or small subset) views, ignoring intra-primitive diversity and global data distr
Approach
Online within-primitive clustering (local-aware prototype assignment via generalized conditional gradient with a coherence regularizer, plus momentum prototype updating) discovers K prototypes per attribute and object; these prototypes anchor a prototype-based contrastive loss and an HSIC-based attribute-object decorrelation loss, and are discarded at test time.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add prototype-based contrastive learning (ablation) | ClusPro without clustering baseline · UT-Zappos CW | AUC: 41.0 -> 45.9 | Improved |
| Add prototype-based decorrelation learning (ablation) | ClusPro without clustering baseline · UT-Zappos CW | AUC: 41.0 -> 43.8 | Improved |
| Add prototype-based contrastive learning (ablation) | ClusPro without clustering baseline · C-GQA CW | AUC: 11.8 -> 14.2 | Improved |
| Add prototype-based decorrelation learning (ablation) | ClusPro without clustering baseline · C-GQA CW | AUC: 11.8 -> 13.7 | Improved |
| Use five prototypes per primitive rather than one (ablation) | ClusPro · UT-Zappos CW | AUC: 42.8 -> 46.6 | Improved |
| Increase prototype count beyond 5 (ablation) | ClusPro · UT-Zappos CW | AUC: 46.6 -> 46.0 (K=10); 46.6 -> 45.9 (K=20) | Hurt |
| Use a high but not extreme prototype momentum coefficient (mu=0.99) (ablation) | ClusPro · UT-Zappos CW | AUC: 42.6 (mu=0) -> 46.6 (mu=0.99) | Improved |
| Remove either attribute or object clustering branch (ablation) | ClusPro · UT-Zappos CW | AUC: 46.6 -> 44.3 (attribute only); 46.6 -> 44.1 (object only) | Hurt |
| Replace local-aware clustering with cosine similarity updating (ablation) | ClusPro · UT-Zappos CW | AUC: 46.6 -> 45.0 | Hurt |
| Replace local-aware clustering with classical optimal transport (ablation) | ClusPro · UT-Zappos CW | AUC: 46.6 -> 45.8 | Hurt |
| Use prototype counts varying from 1 to 5 by primitive sample proportion instead of fixed K=5 (ablation) | ClusPro · UT-Zappos CW | AUC: 46.6 -> 46.8 | Improved |
| Use ClusPro rather than the strongest reported baseline | CLIP ViT-L/14 CZSL · MIT-States CW | AUC: +1.4 | Improved |
Experimental setup
Datasets: MIT-States UT-Zappos C-GQA
Benchmarks: MIT-States UT-Zappos C-GQA
Metrics: Seen accuracy Unseen accuracy HM AUC
Baselines: CLIP CoOp PCVL CSP DFSP GIPCOL CDS-CZSL Troika PLID
Builds on: CLIP CSP DFSP Troika Sinkhorn-Knopp optimal transport
Closest papers by meaning
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.88)
- Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.88)
- Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning (CVPR 2024 · similarity 0.87)
- Compositional Few-Shot Class-Incremental Learning (ICML 2024 · similarity 0.86)
- Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.86)
- ProCC: Progressive Cross-Primitive Compatibility for Open-World Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.86)
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.
