Forecasting evolving clinical risks relies on intrinsic pathological dependencies rather than mere chronological proximity, yet current methods struggle with coarse binary supervision and physical timestamps. To align predictive modeling with clinical logic, we propose the Medical-semantics Aware Time-ALiBi Transformer (MATA-Former), utilizing event semantics to dynamically parameterize attention weights to prioritize causal validity over time lags. Furthermore, we introduce Plateau-Gaussian Soft Labeling (PSL), reformulating binary classification into continuous multi-horizon regression for full-trajectory risk modeling. Evaluated on SIICU -- a newly constructed dataset featuring over 506k events with rigorous expert-verified, fine-grained annotations -- and the MIMIC-IV dataset, our framework demonstrates superior efficacy and robust generalization in capturing risks from text-intensive, irregular clinical time series.
@article{arxiv.2604.01727,
title = {MATA-Former & SIICU: Semantic Aware Temporal Alignment for High-Fidelity ICU Risk Prediction},
author = {Zhichong Zheng and Xiaohang Nie and Xueqi Wang and Yuanjin Zhao and Haitao Zhang and Yichao Tang},
journal= {arXiv preprint arXiv:2604.01727},
year = {2026}
}