Research map / Continual learning / Class-incremental
MgSvF: Multi-Grained Slow versus Fast Framework for Few-Shot Class-Incremental Learning
Hanbin Zhao, Yongjian Fu, Mintong Kang, Qi Chuan Tian, Fei Qing Wu, Xi Li
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 frequency-aware intra-space regularization to the baseline (ablation) | SDC-style NCM baseline · CIFAR100 | last-session accuracy: 34.46 -> 50.60 | Improved |
| add frequency-aware intra-space regularization to the baseline (ablation) | baseline · CUB200 | last-session accuracy: 42.58 -> 49.96 | Improved |
| add frequency-aware intra-space regularization to the baseline (ablation) | baseline · MiniImageNet | last-session accuracy: 41.71 -> 44.46 | Improved |
| add inter-space slow/fast composition (concatenation) to the baseline (ablation) | baseline · CUB200 | last-session accuracy: 42.58 -> 53.16 | Improved |
| add inter-space slow/fast composition (concatenation) to the baseline (ablation) | baseline · CIFAR100 | last-session accuracy: 34.46 -> 49.11 | Improved |
| add inter-space slow/fast composition (concatenation) to the baseline (ablation) | baseline · MiniImageNet | average accuracy: 49.96 -> 51.90 | Improved |
| replace simple concatenation with PCA-reduced composition (ablation) | inter-space composition · CUB200 | average accuracy: 60.68 -> 61.15 | Improved |
| replace simple concatenation with PCA-reduced composition (ablation) | inter-space composition · MiniImageNet | average accuracy: 51.90 -> 53.24 | Improved |
| replace simple concatenation with PCA-reduced composition (ablation) | inter-space composition · CIFAR100 | last-session accuracy: 49.11 -> 50.51 | Improved |
| increase number of frequency groups from 8 to 16 (ablation) | MgSvF intra-space · MiniImageNet | average accuracy: 53.24 -> 52.78 | No clear effect |
| use fewer frequency groups (2 instead of 8) (ablation) | MgSvF intra-space · MiniImageNet | average accuracy: 53.24 -> 52.76 | No clear effect |
| use only slow-updated space (a=0) or only fast-updated space (a=1) instead of the composite (ablation) | inter-space composition · CUB200 | accuracy | Hurt |
Closest papers by meaning
- Few-Shot Class-Incremental Learning via Class-Aware Bilateral Distillation (CVPR 2023 · similarity 0.87)
- Fine-Grained Knowledge Selection and Restoration for Non-exemplar Class Incremental Learning (AAAI 2024 · similarity 0.87)
- Tripartite Weight-Space Ensemble for Few-Shot Class-Incremental Learning (CVPR 2025 · similarity 0.86)
- Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners (CVPR 2024 · similarity 0.85)
- M2SD:Multiple Mixing Self-Distillation for Few-Shot Class-Incremental Learning (AAAI 2024 · similarity 0.85)
- Multi-layer Rehearsal Feature Augmentation for Class-Incremental Learning (ICML 2024 · similarity 0.84)
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.
