AI Paper Atlas

Research map / Detection and segmentation / CNN features

Retriever Encoder Selection Matters for In-Context Learning-based Medical Segmentation

Fan Wang, Zhongyi Han, Yongshun Gong, Yilong Yin

AAAI 2026 · Detection and segmentation › CNN features · approach: Domain-specific segmentation and detection · 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
select RE per query with S3 (IRES) instead of GBC task-level selectionUniverSeg · FUNDUS-OD avgDSC: 72.2 -> 79.4Improved
per-query S3 selection vs oracle per-dataset best encoder (ablation)ICL-MedSAM · FUNDUS-OD avgDSC: 88.4 -> 91.4Improved
per-query S3 selection vs best fixed encoder (ablation)ICL-SAM · FUNDUS-OC avgDSC: IRES 56.9 vs Best-S 53.3 on OC avg; UniverSeg 57.5 vs 59.1Hurt
GBC feature-based encoder selection vs random retrieval (ablation)ICL-SAM · BraTS2020 avgDSC: 17.5 -> 19.2No clear effect
add IRES encoder selectionICL-SAM · BraTS2020 avgDSC: 17.5 -> 32.4Improved
add IRES encoder selection (ablation)ICL-MedSAM · BraTS2020 LGGDSC: 33.1 -> 31.4Hurt
add parallel prediction to per-query S3 selection (ablation)ICL-MedSAM · REFUGEtime (min): 15.1m -> 7.7mImproved
add reciprocal neighbor reuse on top of parallel prediction (ablation)ICL-MedSAM · REFUGEtime (min): 7.7m -> 5.7mImproved
increase reciprocal neighbors m (ablation)IRES · RIM-ONE-r3DSC: up to 5% dropHurt
increase reciprocal neighbors m (ablation)IRES · REFUGE, Drishti-GSDSC: within 1%No clear effect
permute order of top-3 demonstrations (ablation)IRES · RIM-ONE-r3 optic cupDSC: <0.5% variationNo clear effect
add more ICL demonstrations (ablation)ICL-MedSAM · RIM-ONE-r3 optic cupDSC: diminishing returnsImproved

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