Timer-S1 is a released 8.3B-parameter MoE time series model that achieves state-of-the-art MASE and CRPS scores on GIFT-Eval using serial scaling and Serial-Token Prediction.
Strictly proper scoring rules, prediction, and estimation.Journal of the American statistical Association, 102(477):359–378
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
A supervision construction procedure generates explicit support and controlled non-support examples (counterfactual and topic-related negatives) without manual annotation, producing verifiers that demonstrate genuine evidence dependence in radiology tasks.
Empirical comparison of deep ensembles and Monte Carlo dropout with GNLL and MSE losses, plus recalibration, shows DE and recalibrated GNLL perform best for predictive robustness and uncertainty calibration in PPG-based BP estimation under domain shift.
citing papers explorer
-
Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling
Timer-S1 is a released 8.3B-parameter MoE time series model that achieves state-of-the-art MASE and CRPS scores on GIFT-Eval using serial scaling and Serial-Token Prediction.
-
Case-Grounded Evidence Verification: A Framework for Constructing Evidence-Sensitive Supervision
A supervision construction procedure generates explicit support and controlled non-support examples (counterfactual and topic-related negatives) without manual annotation, producing verifiers that demonstrate genuine evidence dependence in radiology tasks.
-
Uncertainty Reliability Under Domain Shift: An Investigation for Data-Driven Blood Pressure Estimation in Photoplethysmography
Empirical comparison of deep ensembles and Monte Carlo dropout with GNLL and MSE losses, plus recalibration, shows DE and recalibrated GNLL perform best for predictive robustness and uncertainty calibration in PPG-based BP estimation under domain shift.