Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
PSVMA+: Exploring Multi-Granularity Semantic-Visual Adaption for Generalized Zero-Shot Learning
Man Liu, Huihui Bai, Feng Li, Chunjie Zhang, Yunchao Wei, Meng Wang, Tat‐Seng Chua, Yao Zhao
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 SMID (semantic-related instance adaption) to the baseline (ablation) | ViT baseline · CUB (GZSL, Proposed Split) | H: 63.8 -> 69.7 | Improved |
| add SMID to the baseline (ablation) | ViT baseline · SUN (GZSL, Proposed Split) | H: 36.0 -> 47.7 | Improved |
| add SMID to the baseline (ablation) | ViT baseline · AwA2 (GZSL, Proposed Split) | H: 67.8 -> 69.5 | Improved |
| add IMSE (instance-centric attributes) on top of SMID (ablation) | baseline + SMID · CUB (GZSL, Proposed Split) | H: 69.7 -> 72.4 | Improved |
| add IMSE on top of SMID (ablation) | baseline + SMID · SUN (GZSL, Proposed Split) | H: 47.7 -> 51.7 | Improved |
| add IMSE on top of SMID (ablation) | baseline + SMID · AwA2 (GZSL, Proposed Split) | H: 69.5 -> 74.6 | Improved |
| add SCGL (selective cross-granularity learning) (ablation) | DSVTM model · CUB (GZSL, Proposed Split) | H: 72.4 -> 73.3 | Improved |
| add SCGL (ablation) | DSVTM model · SUN (GZSL, Proposed Split) | H: 51.7 -> 54.2 | Improved |
| add SCGL (ablation) | DSVTM model · AwA2 (GZSL, Proposed Split) | H: 74.6 -> 76.2 | Improved |
| add AMGF (adaptive multi-granularity fusion) on top of SCGL (ablation) | PSVMA+ · AwA2 (GZSL, Proposed Split) | H: 76.2 -> 79.8 | Improved |
| replace L1 with granularity-aware KL loss in SCGL (ablation) | PSVMA+ · CUB (GZSL, Proposed Split) | H: 72.1 -> 74.6 | Improved |
| use granularity-aware KL loss instead of JSD (ablation) | PSVMA+ · CUB (GZSL, Proposed Split) | H: 72.6 -> 74.6 | Improved |
Later papers that compared with this method
5 later accepted papers report PSVMA as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- When Generalized Zero-Shot Learning Meets PU Learning: A Plug-and-Play Framework for Seen-Class Bias Mitigation (ICML 2026 · better on 2 of 3 shared results)
- SAGE: Structured Attribute-Guided Enhancement for GZSL (AAAI 2026 · better on 3 of 3 shared results)
- Incentivizing Generative Zero-Shot Learning via Outcome-Reward Reinforcement Learning with Visual Cues (CVPR 2026 · better on 1 of 1 shared result)
- Visual-Augmented Dynamic Semantic Prototype for Generative Zero-Shot Learning (CVPR 2024 · better on 2 of 2 shared results)
- Improving Zero-Shot Generalization for CLIP with Variational Adapter (ECCV 2024 · better on 1 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.89)
- Epsilon: Exploring Comprehensive Visual-Semantic Projection for Multi-Label Zero-Shot Learning (AAAI 2025 · similarity 0.88)
- ZeroMamba: Exploring Visual State Space Model for Zero-Shot Learning (AAAI 2025 · similarity 0.88)
- Progressive Semantic-Guided Vision Transformer for Zero-Shot Learning (CVPR 2024 · similarity 0.88)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.88)
- Visual-Augmented Dynamic Semantic Prototype for Generative Zero-Shot Learning (CVPR 2024 · 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.
