Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Semi-Supervised Vocabulary-Informed 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 |
|---|---|---|---|
| Add open-vocabulary maximum-margin constraints to source-class constraints (ablation) | SS-Voc · AwA zero-shot | accuracy (%): 58.6/60.3 -> 59.5/68.4 | Helped |
| Add open-vocabulary maximum-margin constraints to source-class constraints (ablation) | SS-Voc · AwA supervised | accuracy (%): 52.9/58.2 -> 53.6/58.6 | Helped |
| Fine-tune the word vector space with V (ablation) | SS-Voc W · AwA zero-shot | accuracy (%): 59.5/68.4 -> 61.1/68.9 | Helped |
| Fine-tune the word vector space with V (ablation) | SS-Voc W · AwA supervised | accuracy (%): 53.6/58.6 -> 53.9/59.1 | Helped |
| Add open-vocabulary maximum-margin constraints to source-class constraints (ablation) | SS-Voc · ImageNet 2010 zero-shot | accuracy (%): 8.0 -> 8.2 | Helped |
| Fine-tune the word vector space with V (ablation) | SS-Voc W · ImageNet 2010 zero-shot | accuracy (%): 8.2 -> 8.9 | Helped |
| Add open-vocabulary maximum-margin constraints to source-class constraints (ablation) | SS-Voc · ImageNet 2012 supervised | accuracy (%): 34.2 -> 36.3 | Helped |
| Fine-tune the word vector space with V (ablation) | SS-Voc W · ImageNet 2012 supervised | accuracy (%): 36.3 -> 37.1 | Helped |
| Use vocabulary-informed maximum-margin learning instead of SVR-Map | SS-Voc · Open-set recognition on ImageNet vocabulary | top-k accuracy | Mixed |
| Use vocabulary-informed maximum-margin learning instead of SVR-Map | SS-Voc · Open-set supervised-like recognition | accuracy | Helped |
| Increase the number of labeled training instances | SS-Voc · ImageNet supervised recognition | accuracy | Mixed |
Closest papers by meaning
- S3A: Towards Realistic Zero-Shot Classification via Self Structural Semantic Alignment (AAAI 2024 · similarity 0.88)
- Boosting Segment Anything Model Towards Open-Vocabulary Learning (AAAI 2025 · similarity 0.88)
- TPR: Topology-Preserving Reservoirs for Generalized Zero-Shot Learning (NeurIPS 2024 · similarity 0.88)
- Label Propagation for Zero-shot Classification with Vision-Language Models (CVPR 2024 · similarity 0.87)
- Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning (CVPR 2025 · similarity 0.87)
- Epsilon: Exploring Comprehensive Visual-Semantic Projection for Multi-Label Zero-Shot Learning (AAAI 2025 · similarity 0.87)
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.