Research map / Continual learning / Class-incremental
EKPC: Elastic Knowledge Preservation and Compensation for Class-Incremental Learning
Huaijie Wang, De Cheng, Lingfeng He, Yan Li, Jie Li, Nannan Wang, Xinbo Gao
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 IPR (importance-aware parameter regularization) (ablation) | shared adapter baseline · Split-ImageNetA (10 sessions) | A_Last: 57.08 -> 59.14 | Improved |
| add IPR (importance-aware parameter regularization) (ablation) | shared adapter baseline · Split-ImageNetR (10 sessions) | A_Last: 77.71 -> 78.75 | Improved |
| add TSDC (trainable semantic drift compensation) (ablation) | shared adapter baseline · Split-ImageNetA (10 sessions) | A_Last: 57.08 -> 64.10 | Improved |
| add TSDC (trainable semantic drift compensation) (ablation) | shared adapter baseline · Split-DomainNet (10 sessions) | A_Last: 82.75 -> 86.24 | Improved |
| add IPR on top of adapter + TSDC (ablation) | adapter + TSDC · Split-ImageNetR (10 sessions) | A_Last: 80.42 -> 80.60 | Improved |
| add global importance (GI) only (ablation) | shared adapter baseline · Split-ImageNetA (10 sessions) | A_Last: 57.08 -> 58.13 | Improved |
| add local importance (LI) only (ablation) | shared adapter baseline · Split-ImageNetR (10 sessions) | A_Last: 77.71 -> 78.36 | Improved |
| add both GI and LI over GI alone (ablation) | adapter + GI · Split-ImageNetA (10 sessions) | A_Last: 58.13 -> 59.14 | Improved |
| add uniform parameter regularization (PR) (ablation) | shared adapter baseline · Split-ImageNetR (10 sessions) | A_Last: 77.71 -> 74.42 | Hurt |
| replace uniform PR with IPR (ablation) | adapter + regularization · Split-ImageNetR (10 sessions) | A_Last: 74.42 -> 78.75 | Improved |
| replace static drift compensation with trainable TSDC (ablation) | unified classifier training · Split-ImageNetA (10 sessions) | A_Last: 62.43 -> 64.10 | Improved |
| add static drift compensation over no drift compensation (ablation) | unified classifier training · Split-ImageNetA (10 sessions) | A_Last: 59.82 -> 62.43 | Improved |
Closest papers by meaning
- eTag: Class-Incremental Learning via Embedding Distillation and Task-Oriented Generation (AAAI 2024 · similarity 0.86)
- Knowledge Memorization and Rumination for Pre-trained Model-based Class-Incremental Learning (CVPR 2025 · similarity 0.86)
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.86)
- Navigating Semantic Drift in Task-Agnostic Class-Incremental Learning (ICML 2025 · similarity 0.86)
- AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projection (NeurIPS 2025 · similarity 0.86)
- Semantic Shift Estimation via Dual-Projection and Classifier Reconstruction for Exemplar-Free Class-Incremental Learning (ICML 2025 · similarity 0.86)
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.
