AdaPCLA：面向长尾纵向 EHR 生成的自适应先验校准 Logit 调整

This paper has been withdrawn by Shuai Cui

No PDF available, click to view other formats

Abstract:Generative modeling of longitudinal Electronic Health Records is increasingly important for privacy-preserving research, yet standard autoregressive models tend to underrepresent the co-occurrence structure of tail events (i.e., diseases, symptoms), reducing the fidelity and faithfulness of generated data for rare subpopulations. To this end, we propose AdaPCLA framework, which enables generative models to adaptively fit and generate EHR data through a data distribution-aware training strategy; this is achieved by internalizing data knowledge parameters by simulated annealing training. It also supports training-free adaptation to a diverse clinical population for generation through zero-shot distribution control. Moreover, our theoretical analysis characterizes rare-code logit updates through the label-wise empirical NTK and derives a prior-internalization bound for how annealing speed and NTK conditioning affect retained prior signals. Experiments on real-world data show that AdaPCLA achieves consistent gains in tail plausibility, downstream utility, and zero-shot control; in particular, it improves TailPairSeen over HALO by 114.2% on MIMIC-III and 65.1% on MIMIC-IV, outperforms GPT-style generation by 3.5% F1 for zero-shot cross-population adaptation.

Comments:	The authors have identified a critical error/issue in the paper that affects the validity of the results. We are withdrawing to address these issues
Subjects:	Machine Learning (cs.LG)
ACM classes:	I.2.6; J.3
Cite as:	arXiv:2607.12645 [cs.LG]
	(or arXiv:2607.12645v2 [cs.LG] for this version)
	https://doi.org/10.48550/arXiv.2607.12645 arXiv-issued DOI via DataCite

Submission history

From: Shuai Cui [view email]
[v1] Tue, 14 Jul 2026 11:23:51 UTC (4,553 KB)
[v2] Wed, 14 Oct 2026 03:57:37 UTC (1 KB) (withdrawn)