AI Paper Atlas

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

Training-Free Model Merging for Multi-target Domain Adaptation

Wenyi Li, Huan-ang Gao, Mingju Gao, Beiwen Tian, Rong Zhi, Hao Zhao

ECCV 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 15 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
merge buffers (BN statistics) in addition to parameters (ablation)Model merging (ResNet101) · GTA -> Cityscapes+IDDharmonic mean mIoU: 54.8 -> 56.3Improved
merge buffers in addition to parameters (ablation)Model merging (MiT-B5) · GTA -> Cityscapes+IDDharmonic mean mIoU: 60.2 -> 60.8Improved
mid-point parameter merging of independently adapted models (ablation)HRDA STDA models (ResNet101) · GTA -> Cityscapes+IDDharmonic mean mIoU: 52.1 -> 54.8Improved
use Git Re-Basin permutation alignment instead of plain mid-point merging (ablation)Model merging · GTA -> Cityscapes+IDDharmonic mean mIoUNo clear effect
initialize the two models from distinct (not same) weights (ablation)Linear interpolation between adapted models · GTA -> Cityscapes/IDDinterpolated harmonic mean mIoUHurt
same initial weights but without pretraining (ablation)Linear interpolation between adapted models · GTA -> Cityscapes/IDDinterpolated harmonic mean mIoUHurt
disable EMA teacher (copy student weights) in self-training (ablation)Linear mode connectivity · GTA -> Cityscapes/IDDinterpolation performanceNo clear effect
train models on disjoint synthetic subsets (ablation)Linear mode connectivity · synthetic subsets -> Cityscapesinterpolation performanceNo clear effect
merge only backbones, keep separate decoder headsModel merging (ResNet101) · GTA -> Cityscapes+IDDharmonic mean mIoU: 56.3 -> 57.7Improved
merge models from more target domains (four vs two)Model merging · GTA -> C, I, A, Dharmonic mean mIoU: 33.5 -> 36.4 (all four); best 39.3Improved
merge models rather than training on combined data (SYNTHIA, MiT-B5)Model merging · SYNTHIA -> Cityscapes+IDDharmonic mean mIoU*: 62.5 -> 64.8Improved

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