Research map / Continual learning / Class-incremental
CLOSER: Towards Better Representation Learning for Few-Shot Class-Incremental Learning
Junghun Oh, Sungyong Baik, Kyoung Mu Lee
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 inter-class distance minimization (L_inter) on top of low temperature and SSC (ablation) | CLOSER · CIFAR100 | whole-class accuracy A_W (%): 49.98 -> 53.32 | Improved |
| add L_inter on top of low temperature and SSC (ablation) | CLOSER · CIFAR100 | base-class accuracy A_B (%): 66.58 -> 70.72 | Improved |
| add L_inter on top of low temperature and SSC (ablation) | CLOSER · CIFAR100 | new-class accuracy A_N (%): 25.08 -> 27.23 | Improved |
| use lower temperature in softmax cross-entropy (ablation) | baseline · CIFAR100 | new-class accuracy A_N (%): 20.95 -> 24.10 | Improved |
| use lower temperature in softmax cross-entropy (ablation) | baseline · CIFAR100 | base-class accuracy A_B (%): 70.13 -> 68.33 | Hurt |
| add self-supervised contrastive loss L_ssc to baseline (ablation) | baseline · CIFAR100 | new-class accuracy A_N (%): 20.95 -> 22.80 | Improved |
| add L_inter alone without any spreading loss (ablation) | baseline · CIFAR100 | new-class accuracy A_N (%): 20.95 -> 20.05 | Hurt |
| apply CLOSER representation learning instead of baseline | FSCIL with fixed feature extractor · CUB200 | performance drop PD (%): 19.83 -> 15.82 | Improved |
| CLOSER vs ALICE on miniImageNet last-session accuracy | FSCIL · miniImageNet | accuracy (%): 55.70 -> 53.33 | Hurt |
Later papers that compared with this method
2 later accepted papers report CLOSER as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Consistency-Driven Calibration and Matching for Few-Shot Class Incremental Learning (ICLR 2026 · better on 2 of 2 shared results)
- Flexi-FSCIL: Adaptive Knowledge Retention for Breaking the Stability-Plasticity Dilemma in Few-Shot Class-Incremental Learning (ICCV 2025 · better on 1 of 1 shared result)
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
- Few-Shot Class-Incremental Learning from an Open-Set Perspective (ECCV 2022 · similarity 0.91)
- Few-Shot Class-Incremental Learning via Training-Free Prototype Calibration (NeurIPS 2023 · similarity 0.91)
- Few-Shot Class-Incremental Learning via Class-Aware Bilateral Distillation (CVPR 2023 · similarity 0.91)
- Tripartite Weight-Space Ensemble for Few-Shot Class-Incremental Learning (CVPR 2025 · similarity 0.91)
- Adaptive Decision Boundary for Few-Shot Class-Incremental Learning (AAAI 2025 · similarity 0.90)
- OrCo: Towards Better Generalization via Orthogonality and Contrast for Few-Shot Class-Incremental Learning (CVPR 2024 · similarity 0.89)
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.
