Research map / Continual learning / Domain and prompt-based
Non-Exemplar Domain Incremental Learning via Cross-Domain Concept Integration
Qiang Wang, Yuhang He, Songlin Dong, Xinyuan Gao, Shaokun Wang, Yihong Gong
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 |
|---|---|---|---|
| Use a unified classifier plus domain alignment instead of extending classifiers per domain (ablation) | NEDIL on ViT · DomainNet DIL | average accuracy AT: 50.62 -> 61.58 (baseline vs UC+DSA with K-NN) | Improved |
| Add Patch Shuffle Selector instead of K-NN domain selector (ablation) | UC+DSA · DomainNet DIL | average accuracy AT: 61.58 -> 62.23 | Improved |
| Add Patch Shuffle Selector instead of NMC (ablation) | UC+DSA · CORe50 DIL | average accuracy AT: 89.15 -> 90.01 | Improved |
| Add PSS to the S-Prompts baseline (without UC+DSA) (ablation) | S-Prompts baseline · DomainNet DIL | average accuracy AT: 50.62 -> 51.29 | Improved |
| Use CLIP backbone with PINA instead of beating MoP-CLIP on CORe50 | PINA-D (CLIP) · CORe50 DIL | average accuracy AT: 89.23 vs 92.29 (MoP-CLIP) | Hurt |
| Use lowest-quality domain (quickdraw) as the base domain instead of real (ablation) | PINA-D · DomainNet DIL | average accuracy AT: 66.67 -> 56.52 | Hurt |
| Use deep branch alignment (PINA-D) instead of shallow branch at same parameter count (ablation) | DSA module · DomainNet DIL | average accuracy AT: +3.2% | Improved |
| Increase prompt token length to 10 or more (ablation) | PINA · DomainNet DIL | average accuracy AT | Improved |
| Use PINA instead of S-iPrompts at fewer parameters | NEDIL on ViT · DomainNet DIL | average accuracy AT: 50.62 -> 54.86 with 0.6 vs 1.8 extra parameters | Improved |
Closest papers by meaning
- Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models (NeurIPS 2024 · similarity 0.87)
- Dual Consolidation for Pre-Trained Model-Based Domain-Incremental Learning (CVPR 2025 · similarity 0.86)
- Addressing Imbalanced Domain-Incremental Learning through Dual-Balance Collaborative Experts (ICML 2025 · similarity 0.86)
- Componential Prompt-Knowledge Alignment for Domain Incremental Learning (ICML 2025 · similarity 0.86)
- Gradual Divergence for Seamless Adaptation: A Novel Domain Incremental Learning Method (ICML 2024 · similarity 0.86)
- Make Domain Shift a Catastrophic Forgetting Alleviator in Class-Incremental Learning (AAAI 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.
