AI Paper Atlas

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

Robust Nearest Neighbors for Source-Free Domain Adaptation under Class Distribution Shift

Antonio Tejero-de-Pablos, Riku Togashi, Mayu Otani, Shin'ichi Satoh

ECCV 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 1 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 robust nearest neighbors with a Swin-B generic modelPL guided · VisDA-C RSUT (SFDA)class-wise mean accuracy: 83.59 -> 88.84Improved
add robust nearest neighbors with a ResNet generic model on Office-HomePL guided · Office-Home RSUT (SFDA)class-wise mean accuracy: 61.05 -> 59.67Hurt
add robust nearest neighbors with a ResNet50 generic model on DomainNetPL guided · DomainNet subset (SFDA)class-wise mean accuracy: 80.12 -> 72.74Hurt
add robust nearest neighbors with ViT-B on DomainNetPL guided · DomainNet subset (SFDA)class-wise mean accuracy: 80.12 -> 83.9Improved
use Swin-B generic model instead of ViT-B on DomainNet (ablation)PL guided + Ours · DomainNet subset (SFDA)class-wise mean accuracy: 83.9 -> 81.4Hurt
add robust nearest neighbors with Swin-B to basic pseudolabelingPL base · VisDA-C RSUT (SFDA)class-wise mean accuracy: 81.01 -> 86.64Improved
set common-neighbor range K to 100 rather than 10 or 500 (ablation)Robust NN · VisDA-C RSUTclass-wise mean accuracy: 83.96 (K=10), 83.56 (K=500) -> 86.6 (K=100)Improved
add robust nearest neighbors to test-time adaptation pseudolabeling with Swin-BPseudolabel TTA · Office-Home RSUT (TTA)class-wise mean accuracy: 52.34 -> 60.16Improved
add robust nearest neighbors to TTA on VisDA-C versus label shift adapterPseudolabel TTA · VisDA-C RSUT (TTA)class-wise mean accuracy: 47.12 -> 52.49 (vs 72.97 for label shift adapter)Mixed
increase class distribution shift severity (ablation)PL guided + Ours · VisDA-C RSUTclass-wise mean accuracy: ViT-B: 86.72 -> 83.88; ResNet101: 86.60 -> 80.46Hurt

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