Research map / Vision-language and multimodal / CLIP and zero-shot
Collect and Select: Semantic Alignment Metric Learning for Few-Shot Learning
Fusheng Hao, Fengxiang He, Jun Cheng, Lei Wang, Jianzhong Cao, Dacheng Tao
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 attention to the cosine metric (ablation) | SAML · miniImageNet 5-way 1-shot | accuracy: 49.52% -> 56.68% | Improved |
| add attention to the cosine metric (ablation) | SAML · miniImageNet 5-way 5-shot | accuracy: 62.82% -> 71.34% | Improved |
| add attention to the Gaussian metric (ablation) | SAML · miniImageNet 5-way 1-shot | accuracy: 52.35% -> 56.40% | Improved |
| add attention to the Gaussian metric (ablation) | SAML · miniImageNet 5-way 5-shot | accuracy: 68.54% -> 71.28% | Improved |
| increase input size from 84x84 to 224x224 (ablation) | SAML · miniImageNet 5-way 1-shot | accuracy: 52.22% -> 56.68% | Improved |
| increase input size from 84x84 to 224x224 (ablation) | SAML · miniImageNet 5-way 5-shot | accuracy: 66.49% -> 71.34% | Improved |
| increase input size from 84x84 to 224x224 (ablation) | RelationNet · miniImageNet 5-way 1-shot | accuracy: 50.44% -> 50.16% | No clear effect |
| increase input size from 84x84 to 224x224 (ablation) | RelationNet · miniImageNet 5-way 5-shot | accuracy: 65.32% -> 65.98% | Improved |
| increase number of local receptive fields up to 100 (ablation) | SAML · miniImageNet | accuracy | Improved |
| add inception operator for multi-scale features (ablation) | SAML · miniImageNet 5-way 1-shot | accuracy: 56.68% -> 57.69% | Improved |
| add inception operator for multi-scale features (ablation) | SAML · miniImageNet 5-way 5-shot | accuracy: 71.34% -> 73.03% | Improved |
| add inception operator for multi-scale features (ablation) | SAML · CUB 5-way 1-shot | accuracy: 69.33% -> 69.35% | No clear effect |
Closest papers by meaning
- Simple Semantic-Aided Few-Shot Learning (CVPR 2024 · similarity 0.89)
- S3A: Towards Realistic Zero-Shot Classification via Self Structural Semantic Alignment (AAAI 2024 · similarity 0.87)
- VT-FSL: Bridging Vision and Text with LLMs for Few-Shot Learning (NeurIPS 2025 · similarity 0.87)
- Making Large Vision Language Models to Be Good Few-Shot Learners (AAAI 2025 · similarity 0.86)
- DH-Set: Improving Vision-Language Alignment with Diverse and Hybrid Set-Embeddings Learning (CVPR 2025 · similarity 0.86)
- Few-Shot Learning from Gigapixel Images via Hierarchical Vision-Language Alignment and Modeling (NeurIPS 2025 · 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.
