Research map / Continual learning / Class-incremental
Exemplar-Free Class Incremental Learning via Preserving Class-Discriminative Structure
Xin Zhang, Liang Bai, Guanchao Wang, Xian Yang
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 SCC (structural consistency constraint) (ablation) | baseline without SCC/APR · ImageNet-A (10 tasks) | Last-Acc: 60.08 -> 63.60 | Improved |
| add APR (adaptive prototype rectification) (ablation) | baseline without SCC/APR · ImageNet-A (10 tasks) | Last-Acc: 60.08 -> 61.97 | Improved |
| add SCC (ablation) | baseline without SCC/APR · Cars-196 (10 tasks) | Last-Acc: 67.90 -> 74.86 | Improved |
| add APR (ablation) | baseline without SCC/APR · Cars-196 (10 tasks) | Last-Acc: 67.90 -> 77.02 | Improved |
| combine SCC and APR (ablation) | SCC-only model · Cars-196 (10 tasks) | Last-Acc: 74.86 -> 80.23 | Improved |
| replace SDC with APR for prototype rectification (ablation) | SCC-only model · Cars-196 (10 tasks) | Last-Acc: 76.26 -> 80.23 | Improved |
| replace SDC with APR (ablation) | SCC-only model · ImageNet-A (10 tasks) | Last-Acc: 63.88 -> 64.58 | Improved |
| add logit distillation to APR-only model (ablation) | APR-only model · Cars-196 (10 tasks) | Last-Acc: 77.02 -> 67.76 | Hurt |
| add feature distillation to APR-only model (ablation) | APR-only model · Cars-196 (10 tasks) | Last-Acc: 77.02 -> 75.37 | Hurt |
| add logit or feature distillation to APR-only model (ablation) | APR-only model · ImageNet-A (10 tasks) | Last-Acc: 61.97 -> 62.17 (logit), 61.90 (feature) | No clear effect |
| increase covariance correction rho from 0 to 10 (ablation) | APR · Cars-196 (10 tasks) | Last-Acc: 78.49 -> 80.23 | Improved |
| increase covariance correction rho from 0 to 10 (ablation) | APR · ImageNet-A (10 tasks) | Last-Acc: 63.99 -> 64.58 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SLCA — SLCA: Slow Learner with Classifier Alignment for Continual Learning on a Pre-trained Model (ICCV 2023) · better on 2 of 3 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
- Defying Imbalanced Forgetting in Class Incremental Learning (AAAI 2024 · similarity 0.90)
- Make Domain Shift a Catastrophic Forgetting Alleviator in Class-Incremental Learning (AAAI 2025 · similarity 0.90)
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.89)
- Task Confusion and Catastrophic Forgetting in Class-Incremental Learning: A Mathematical Framework for Discriminative and Generative Modelings (NeurIPS 2024 · similarity 0.89)
- Federated Class-Incremental Learning: A Hybrid Approach Using Latent Exemplars and Data-Free Techniques to Address Local and Global Forgetting (ICLR 2025 · similarity 0.88)
- Non-exemplar Online Class-Incremental Continual Learning via Dual-Prototype Self-Augment and Refinement (AAAI 2024 · similarity 0.88)
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.
