AI Paper Atlas

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

Exploring Accurate and Transparent Domain Adaptation in Predictive Healthcare via Concept-Grounded Orthogonal Inference

Pengfei Hu, Chang Lu, Feifan Liu, Yue Ning

ICML 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment · 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 SAE reconstruction and domain classification losses with orthogonal inference (ablation)MMD-aligned encoder · eICU diagnosisweighted F1: 64.87 -> 68.61Improved
add orthogonal inference (W, z) and domain classification loss (ablation)ExtraCare · eICU diagnosisweighted F1: 66.73 -> 68.61Improved
use dictionary-induced metric M instead of Euclidean inner product (ablation)ExtraCare · eICU diagnosisweighted F1: 67.58 -> 68.61Improved
supervise covariate residual with domain classification loss rather than adversarial loss (ablation)ExtraCare · OCHIN diagnosisweighted F1: 71.28 -> 74.05Improved
add M-metric orthogonal residual on OCHIN heart failure (ablation)ExtraCare · OCHIN heart failureAUROC: 95.07 -> 95.48No clear effect
ExtraCare vs best baseline on eICU heart failure F1BUA · eICU heart failureF1: 79.83 -> 79.64No clear effect
apply ExtraCare relative to source-only modelBase · OCHIN diagnosisweighted F1: 63.77 -> 74.05Improved
use simple DANN or DAL alignmentBase · eICU diagnosisweighted F1: 61.09 -> 62.14 (DANN), 61.58 (DAL)No clear effect
use self-training (CST, SSRT) for adaptationBase · eICU and OCHINvariousMixed
use ExtraCare sparse code explanations instead of regular SAEregular SAE · OCHIN facility adaptationLabel-Axis Difference (%): 36 -> 47Improved
use ExtraCare explanations instead of regular SAEregular SAE · OCHIN facility adaptationDomain-Axis Difference (%): 40 -> 42No clear effect
use ExtraCare explanations instead of regular SAEregular SAE · OCHIN facility adaptationQuadrant Calibration Error (%), lower better: 37 -> 32Improved

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