Research map / Continual learning / Domain and prompt-based
HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning
Yu Feng, Zhen Tian, Haoran Luo, Xie Yu, Diancheng Cheng, Haoyue Zheng, Shuai Lyu, Ping Zong, Lianyuan Li, xin ge, Yifan Zhu
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 energy factor (with regularization) to distance-only selection (ablation) | HEDP · CDDB-Hard unknown | average accuracy: 75.8 -> 83.74 | Improved |
| add distance factor to energy-regularized selection (ablation) | HEDP · CDDB-Hard unknown | average accuracy: 81.52 -> 83.74 | Improved |
| add energy regularization loss (boundary + midline) (ablation) | energy-only HEDP · CDDB-Hard known | average accuracy: 78.83 -> 92.75 | Improved |
| add boundary loss only (ablation) | energy-only HEDP · CDDB-Hard known | average accuracy: 78.83 -> 86.85 | Improved |
| add midline loss only (ablation) | energy-only HEDP · CDDB-Hard known | average accuracy: 78.83 -> 91.63 | Improved |
| combine boundary and midline losses (ablation) | energy-only HEDP · CDDB-Hard unknown | average accuracy: 79.22 -> 81.52 | Improved |
| add energy regularization to energy-only selection (ablation) | energy factor · DomainNet known(all) | average accuracy: 66.84 -> 72.54 | Improved |
| add hybrid energy+distance factors (ablation) | energy-regularized HEDP · CORe50 unknown | average accuracy: 94.07 -> 94.66 | Improved |
| vary number of K-means clusters K (ablation) | distance factor · CDDB, DomainNet, CORe50 unknown | average accuracy | No clear effect |
| set alpha or beta to extreme 0.1 (ablation) | hybrid factor weights · unknown domains | average accuracy: decrease of approximately 2-3 percentage points | Hurt |
| place energy boundary Theta too close to or too far from midline Delta (ablation) | energy regularization · unknown domains | average accuracy | Hurt |
Closest papers by meaning
- Boosting Domain Incremental Learning: Selecting the Optimal Parameters is All You Need (CVPR 2025 · similarity 0.88)
- 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)
- Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation (NeurIPS 2025 · similarity 0.86)
- GraphKeeper: Graph Domain-Incremental Learning via Knowledge Disentanglement and Preservation (NeurIPS 2025 · similarity 0.86)
- Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models (NeurIPS 2024 · similarity 0.85)
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.
