AI Paper Atlas

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

sleep2vec: Unified Cross-Modal Alignment for Heterogeneous Nocturnal Biosignals

Weixuan Yuan, Zengrui Jin, Yichen Wang, Donglin Xie, Ziyi Ye, Chao Zhang, Xuesong Chen

ICLR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer · 2 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
use sleep2vec instead of SleepFM foundation modelsleep2vec · SHHS ECG & ABDaccuracy (%): 77.9 -> 82.7Improved
foundation model vs specialized EEG-only modelssleep2vec · SHHS EEG onlyaccuracy (%): 87.7 -> 87.4No clear effect
add metadata-aware weighting (DASH-InfoNCE) instead of vanilla InfoNCE (ablation)sleep2vec · APPLES unseen cohort, full channelsaccuracy (%): 76.8 -> 78.4Improved
DASH-InfoNCE instead of vanilla InfoNCE (ablation)sleep2vec · SHHS full channelsaccuracy (%): 88.4 -> 88.6Improved
DASH-InfoNCE instead of vanilla InfoNCE (ablation)sleep2vec · pre-training retrievalretrieval recall@1: 0.351 -> 0.368Improved
gating fusion instead of mean fusion (ablation)sleep2vec · APPLES unseen cohortaccuracy (%): 78.1 -> 78.4Improved
gating fusion instead of concatenation (ablation)sleep2vec · APPLES unseen cohortaccuracy (%): 76.9 -> 78.4Improved
add 15% masking during pre-training (ablation)sleep2vec · APPLES unseen cohortmacro-F1 (%): 64.0 -> 65.2Improved
raise masking ratio from 15% to 30% (ablation)sleep2vec · APPLES unseen cohortmacro-F1 (%): 65.2 -> 63.4Hurt
increase pre-training data from 25% to 100% (ablation)sleep2vec · SHHS sleep stagingaccuracy (%): 87.1 -> 88.6Improved
scale model from 63.5M to 238.2M parameters (ablation)sleep2vec · SHHS and WSC sleep stagingaccuracyImproved
increase number of input modalities (ablation)sleep2vec · SHHS disease predictionROC-AUCImproved

Earlier methods this paper compares with

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