AI Paper Atlas

Research map / Continual learning / Domain and prompt-based

DK-DDIL: Adaptive Knowledge Retention for Dynamic Domain-Incremental Learning in Medical Imaging

Yuxi Ma, Sujie Liu, Jing Yang, Jiacheng Wang, Yiping Chen, Baptiste Magnier, Liansheng Wang

CVPR 2026 · Continual learning › Domain and prompt-based · approach: Domain-incremental learning with memory · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
add DAM (dynamic rank LoRA) to classifier-only fine-tuning (ablation)FT · Skin Pathology Diagnosisaverage accuracy A: 52.96 -> 75.56Improved
add DCL to classifier-only fine-tuning (ablation)FT · Skin Pathology Diagnosisaverage accuracy A: 52.96 -> 52.98No clear effect
add DCL on top of DAM (ablation)FT+DAM · Skin Pathology Diagnosisfinal accuracy AT: 68.68 -> 70.88Improved
add MKI on top of DAM (ablation)FT+DAM · Skin Pathology Diagnosisaverage accuracy A: 75.56 -> 75.57No clear effect
add MKI and DCL together on top of DAM (ablation)FT+DAM · Skin Pathology Diagnosisfinal accuracy AT: 68.68 -> 71.52Improved
insert DAM into all layers instead of even layers only (ablation)DK-DDIL · Skin Pathology Diagnosisaverage accuracy A: 73.02 -> 77.03Improved
insert DAM into odd layers only instead of all layers (ablation)DK-DDIL · Skin Pathology Diagnosisaverage accuracy A: 77.03 -> 76.97No clear effect
apply DAM to all projections instead of Q only (ablation)DK-DDIL · Skin Pathology Diagnosisaverage accuracy A: 72.23 -> 77.03Improved
raise alpha_init from 0.1 to 0.3 (ablation)DK-DDIL · Skin Pathology Diagnosisaverage accuracy A: 77.03 -> 77.71Mixed
DK-DDIL vs best baseline on OfficeHomeDomain-incremental learning · OfficeHomeaverage accuracy A: 82.22 -> 84.35Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most 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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy