AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Do We Need Perfect Data? Leveraging Noise for Domain Generalized Segmentation

Taeyeong Kim, SeungJoon Lee, Jung Uk Kim, MyeongAh Cho

AAAI 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation · 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 GAP to DGInStyle baseline (ablation)DAFormer + DGInStyle · Avg2 (ACDC, Dark Zurich)mIoU: 34.81 -> 36.33Improved
add UBE to DGInStyle baseline (ablation)DAFormer + DGInStyle · Avg2 (ACDC, Dark Zurich)mIoU: 34.81 -> 35.05Improved
add UBE on top of GAP (ablation)DAFormer + DGInStyle + GAP · Avg2 (ACDC, Dark Zurich)mIoU: 36.33 -> 36.15No clear effect
add UBE on top of GAP (ablation)DAFormer + DGInStyle + GAP · Avg3 (CS, BDD, MV)mIoU: 55.26 -> 56.07Improved
add HAS on top of GAP+UBE (ablation)FLEX-Seg · Avg2 (ACDC, Dark Zurich)mIoU: 36.15 -> 38.04Improved
add HAS on top of GAP+UBE (ablation)FLEX-Seg · Avg3 (CS, BDD, MV)mIoU: 56.07 -> 55.61Hurt
add FLEX-Seg to DGInStyle (DAFormer)DAFormer + DGInStyle · Dark ZurichmIoU: 25.58 -> 29.51Improved
add FLEX-Seg to DGInStyle (HRDA MiT-B5)HRDA + DGInStyle · Mapillary VistasmIoU: 62.47 -> 61.71Hurt
add FLEX-Seg to DGInStyle (HRDA MiT-B5)HRDA + DGInStyle · Average of 5 datasetsmIoU: 48.99 -> 50.07Improved

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