Scenario

Risk Stratification

Single time-independent risk score or survival time

Survival Curve Estimation

Full time-dependent survival probability curves

Metric

Harrell's C-Index

Uno's C-Index

Cumulative/Dynamic AUC

Dataset Size

Leaderboard Metric

Classic

Deep Learning

Foundation Model

Model	Elo	Improvability	Average Rank	MRR
Coxnet (Tuned) classic	1124.9 [1070.8, 1187.7]	0.056 [0.038, 0.072]	4.58	0.343
PopSICL foundation	1082.0 [1015.5, 1162.1]	0.072 [0.044, 0.107]	5.28	0.375
Coxnet classic	1057.7 [1003.7, 1116.6]	0.073 [0.048, 0.095]	5.69	0.341
SSVM (Tuned) classic	1049.7 [974.8, 1127.6]	0.073 [0.053, 0.099]	5.83	0.291
DeepSurv deep	1012.3 [953.0, 1079.3]	0.089 [0.064, 0.114]	6.50	0.236
RSF classic	1000.0 [957.3, 1068.2]	0.080 [0.050, 0.112]	6.72	0.222
RankDeepSurv deep	998.5 [943.5, 1056.1]	0.090 [0.063, 0.119]	6.75	0.216
RSF (Tuned) classic	965.0 [909.5, 1024.7]	0.089 [0.062, 0.118]	7.36	0.193
GBSE (Tuned) classic	965.0 [902.9, 1040.9]	0.088 [0.065, 0.110]	7.36	0.265
SSVM classic	888.7 [821.9, 954.6]	0.131 [0.106, 0.159]	8.75	0.190
GBSE classic	879.3 [796.2, 959.4]	0.110 [0.081, 0.140]	8.92	0.160
TabPFN* foundation	855.5 [769.6, 947.3]	0.152 [0.100, 0.199]	9.33	0.188
DeepHit deep	809.1 [741.7, 881.8]	0.175 [0.124, 0.227]	10.11	0.138
DeepWeiSurv (p=2) deep	556.0 [439.2, 631.9]	0.241 [0.189, 0.289]	13.28	0.079
DeepWeiSurv (p=1) deep	526.7 [401.7, 619.5]	0.251 [0.201, 0.296]	13.53	0.080