AI Paper Atlas

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

Mix and Reason: Reasoning over Semantic Topology with Data Mixing for Domain Generalization

Chaoqi Chen, Luyao Tang, Feng Liu, Gangming Zhao, Yue Huang, Yizhou Yu

NeurIPS 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Foundation-model domain generalization · 20 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 CDM to MiRe (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 65.5Improved
use the class activation map in CDM (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 65.0Improved
use the domain activation map in CDM (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 66.4Improved
apply Gaussian blur to CDM background images (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 65.8Improved
use the domain activation map rather than its inverse (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 65.0Improved
add ASTR to MiRe (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 64.4Improved
enforce cross-domain invariance in ASTR (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 66.2Improved
enforce cross-model invariance in ASTR (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 66.4Improved
model semantic topology with graph structure (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 65.3Improved
aggregate features using semantic anchors (ablation)MiRe · Four-benchmark averageaccuracy: 67.0 vs 66.1Improved
apply CDM to ERER · PACS averageaccuracy: 82.5 (+1.1)Improved
apply CDM to MixStyleMixStyle · PACS averageaccuracy: 83.7 (+0.8)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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