Research map / Probabilistic and scientific modeling / Molecules and proteins
Explicitly Modeling Censoring Produces Superior Survival Predictors
Shi-ang Qi, Yakun Yu, Russell Greiner
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add decomposed event/censoring/confounder representations (SALaD) to deep survival baselines | DeepSurv, N-MTLR, AFTNN-Weibull, AFTNN-LogLogistic · 10 survival datasets | CI, IBS, MAE, D-cal: 109 wins, 28 ties, 19 losses of 156 comparisons | Improved |
| add explicit disentanglement (SALaD) on top of the two-branch 2B extension (ablation) | 2B · semi-SUPPORT, HFCR, SUPPORT | CI, IBS, MAE | Improved |
| use simple two-branch (2B) censoring modeling instead of event-only baselines (ablation) | Deep survival baselines · semi-SUPPORT, HFCR, SUPPORT | CI, IBS, MAE | Improved |
| use 2B instead of SALaD for calibration (ablation) | SALaD · Some datasets | IBS, D-cal | Mixed |
| add distance-based invariance regularizers (beta2 > 0) (ablation) | SALaD · Survival datasets | CI, IBS, MAE | Improved |
| add orthogonal regularizer (beta1 > 0) (ablation) | SALaD · Survival datasets | CI, IBS, MAE | Improved |
| use adversarial training instead of discrepancy-based invariance regularizers (ablation) | SALaD · Survival datasets | performance | Hurt |
Closest papers by meaning
- Deep Copula-Based Survival Analysis for Dependent Censoring with Identifiability Guarantees (AAAI 2024 · similarity 0.91)
- Learning Survival Models with Right-Censored Reporting Delays (NeurIPS 2026 · similarity 0.89)
- Censor Dependent Variational Inference (ICML 2025 · similarity 0.88)
- SurvivalPFN: Amortizing Survival Prediction via In-Context Bayesian Inference (NeurIPS 2026 · similarity 0.88)
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data (NeurIPS 2025 · similarity 0.88)
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring (NeurIPS 2026 · similarity 0.87)
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.
