AI Paper Atlas

Research map / Time series / Causal and generative time series

Reliable Probabilistic Forecasting of Irregular Time Series through Marginalization-Consistent Flows

Vijaya Krishna Yalavarthi, Randolf Scholz, Christian Klötergens, Kiran Madhusudhanan, Stefan Born, Lars Schmidt-Thieme

ICLR 2026 · Time series › Causal and generative time series · approach: Probabilistic and conformal forecasting

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 changedWhereMeasuredEffect
add normalizing flows to the mixture of Gaussians (ablation)MOSES · PhysioNet'12njNLL: 1.063 -> -0.491Improved
add full low-rank covariance to the base Gaussian (ablation)MOSES · PhysioNet'12njNLL: -0.308 -> -0.491Improved
learn mixture weights instead of fixing them to 1/D (ablation)MOSES · PhysioNet'12njNLL: -0.451 -> -0.491Improved
use a mixture of D components instead of one component (ablation)MOSES · PhysioNet'12njNLL: -0.493 -> -0.491No clear effect
use a mixture of D components instead of one component (ablation)MOSES · MIMIC-IIInjNLL: 0.173 -> -0.305Improved
use a mixture of D components instead of one component (ablation)MOSES · USHCNnjNLL: -3.255 -> -3.357Improved
use MOSES (marginalization consistent) instead of ProFITiProFITi · MIMIC-IIImNLL: 0.408 -> 0.163Improved
use MOSES instead of ProFITiProFITi · PhysioNet'12njNLL: -0.647 -> -0.491Hurt
use MOSES instead of ProFITi on long horizon PhysioNet'12 (obs 12h, forecast 36h)ProFITi · PhysioNet'12 12/36mNLL: 2.977 -> 0.040Improved
replace the MOSES encoder with GraFITi (ablation)MOSES · MIMIC-IVnjNLL: -1.668 -> -2.197Improved
replace the MOSES encoder with GraFITi (ablation)MOSES · MIMIC-IVmNLL: -0.634 -> -0.299Hurt
use MOSES instead of ProFITi for CRPSProFITi · MIMIC-IVCRPS: 0.279 -> 0.245Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy