AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Adapting Noise to Data: Generative Flows from Learned 1D Processes

Jannis Chemseddine, Gregor Kornhardt, Richard Duong, Gabriele Steidl

ICML 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Flow matching and flow sampling

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
learn per-coordinate quantile noise instead of Gaussian noiseOT flow matching · HRRR-mini precipitationextreme event magnitude error: 0.2455 -> 0.0634Improved
learn per-coordinate quantile noise instead of hand-tuned Student-t noiseOT flow matching · HRRR-mini precipitationtail KS distance: 0.1014 -> 0.0393Improved
replace Gaussian noise with heavy-tailed noise (Student-t) (ablation)flow matching · HRRR-mini precipitationkurtosis deviation: 4.930 -> 2.890Improved
learn quantile noise instead of Gaussian on a larger modelOT flow matching, 55M U-Net · CIFAR-10FID: 3.37 -> 3.25Improved
learn quantile noise instead of Gaussian on 35M U-NetOT flow matching, 35M U-Net · CIFAR-10FIDNo clear effect
learned quantile noise at reduced U-Net capacityOT flow matching · MNIST (channels 8/16/32)FIDImproved
use learned data-adapted latent instead of Gaussian with OT couplingOT flow matching · checkerboardtransport path lengthImproved
use learned latent for heavy-tailed funnel instead of Gaussian or compact noiseflow matching, small network · Neal's funnelsample quality (qualitative)Improved
add stop-gradient on target displacement for the quantile (ablation)quantile network training · joint trainingcollapse avoidanceImproved

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