AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning

Hao Zhou, Simon A. Lee, Cyrus Tanade, Keum San Chun, Juhyeon Lee, Migyeong Gwak, Megha Thukral, Justin Sung, Eugene Hwang, Mehrab Bin Morshed, Li Zhu, Viswam Nathan and 3 more

ICML 2026 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer · 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
pretrain PPG encoder with xMAE masked cross-modal ECG reconstruction instead of unimodal or exchangeable-view multimodal baselinesPPG encoder · 19 downstream tasks (linear probing)AUROC / MAE: 15 of 19 tasks bestImproved
xMAE vs strongest baseline on hypertension (lab)PPG encoder · Hypertension (lab)AUROC: 56.8 -> 68.8Improved
xMAE vs strongest baseline on A1CPPG encoder · A1CAUROC: 52.5 -> 65.1Improved
xMAE on BMI (lab) and Deep sleep, where baselines were betterPPG encoder · BMI (lab); sleep DeepMAE; AUROC: BMI 4.09 vs 3.79 (PaPaGei-P); Deep 55.9 vs 57.5Mixed
use continuous block ECG masking instead of random masking (ablation)xMAE · Hypertension (lab) and Ectopic BeatsAUROC: 63.0 -> 68.8; 84.4 -> 87.8Improved
use curriculum masking ratio instead of a fixed ratio (ablation)xMAE · Hypertension (lab) and Ectopic BeatsAUROC: 66.9 -> 68.8; 85.8 -> 87.8Improved
use cross-modal reconstruction instead of multimodal self-reconstruction (ablation)xMAE · Hypertension (lab) and Ectopic BeatsAUROC: 65.8 -> 68.8; 83.4 -> 87.8Improved
remove directional cross-attention (ablation)xMAE · ECG reconstruction errormean absolute reconstruction errorHurt
xMAE pretraining vs multimodal MAE baseline for ECG-PPG timingmultimodal MAE · ECG R-peak to PPG onset delaymedian delay error: 21.5 ms median, 53.3% reductionImproved
xMAE vs handcrafted PPG features (all and timing-only)logistic regression on features · Hypertension (free-living)AUROC: 58.4 -> 58.5No clear effect
xMAE (smaller model and less data) vs open-weight foundation modelsopen-source PPG models · Ectopic beatsAUROC: AnyPPG 89.3 vs xMAE 87.8Mixed

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