Survival Analysis
Overview
Survival analysis estimates when an event may occur and how risk changes over time, even when follow-up ends before the event is observed. Methods range from Kaplan–Meier curves, Cox and accelerated failure-time models to random survival forests, neural hazard models, and newer in-context models pretrained on synthetic data. Current work applies these tools to longitudinal, multimodal, and clinical data, including competing risks, personalized treatment decisions, and causal effects. A central challenge is ensuring predictions are not only good at ranking risk but also calibrated, fair, and reliable when censoring or data distributions differ across settings.
Latest papers
110 papers6 papers in the last four weeks, down 50% on the four weeks before. 0.0% of all new papers.
K-Survival Means
A Multi-Cohort Validation of Censoring-Aware Conformal Lower Predictive Bounds for Pathology Survival Models
The C-index illusion: discrimination without calibration in published survival models
Censoring-Aware In-Context Learning for Generalized Supplier Lead Time Estimation in Supply Chain Planning
SurvCF(t): Counterfactual Explanations for Survival Analysis in Predictive Maintenance Multivariate Time Series Data
Value-Monotonicity Matters: A Concordance Loss for Deep Survival Prediction
Multimodal Empirical Bayes Variational Autoencoders for Joint Longitudinal and Time-to-Event Modeling
A Temporal Machine Learning-Based Time-to-Event Model for Predicting ALS Progression and Healthcare Utilization
Pitfalls of Administrative Censoring in Survival Models with Time-Indexed Inputs
CT-CLIP Representations for Multimodal Lung Cancer Survival Prediction
MergeSurv: Merging-Based Continual Learning for Survival Analysis on Whole-Slide Images
SHIFT: Survival Prediction from Incomplete and Heterogeneous Genomic Data
A Filtered Mixture-of-Generators for Fully Synthetic Survival Training
Dynamic Prediction of Alternating Recurrent Events via Neural Network
Federated Survival Analysis in Healthcare: A Multi-Model Evaluation on Cross-Institutional Heterogeneous Breast Cancer Data
SSH-Net: A Deep Neural Network for Predicting Failure Time Distribution Functions under Competing Risks with Application to GPU Data
Evidential Fusion Network for Multimodal Survival Prediction under Missing Modalities
ChronoSurv: A Clinical Pathway-Guided Graph Framework for Multimodal Survival Analysis
iLENS: Interpretable LLM-Guided Mixture-of-Experts for Neuroimaging Survival Analysis
Expert-Driven Survival Machines: Improving Stratification and Interpretability in Multiple Clinical Cohorts
Time-Conditioned and Multi-Time Survival Prediction from 2D PET/CT Projections in Lung Cancer
Tabular Foundation Models for Clinical Survival Analysis via Survival-Aware Adaptation
Convergence Rates for Neural-Network Estimation with Current-Status Data
A Guide to Estimating Conditional Average Treatment Effects in Competing Risks Settings
GD-MIL: Grade-Disentangled Multiple Instance Learning for Multimodal Biochemical Recurrence Prediction in Prostate Cancer
From Hazard Functions to Language Space: Cox-Supervised Distillation of Survival Risk into a Large Language Model
MEC-Cox: Machine-Learning-Assisted Generalized Entropy Calibration for ATT Marginal Hazard-Ratio Estimation
Proper Scoring Rules for Right-Censored Survival Data
SurvPFN: Towards Foundation Models for Survival Predictions
SurvPFN, a prior-data fitted network (PFN), for survival prediction tasks. We pretrain SurvPFN on millions of synthetic survival prediction tasks to learn survival via distributional regression that accounts for censored data. SurvPFN works by (1) generating data with Weibull event times and a non-informative censoring mechanism; (2) integrating a censored event indicator; and (3) minimizing a censored negative log-likelihood. On SurvSet, a collection of real-world survival tasks, SurvPFN is highly competitive with classical and deep survival baselines without per-dataset fitting, a survival-specific architecture, or feature engineering. We show that survival can be treated as a continuous-time distributional regression problem with censored loss, unlocking the power of PFNs for time-to-event predictions.