AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Don't Forget Its Variance! The Minimum Path Variance Principle for Accurate and Stable Score-Based Models

Wei Chen, Jiacheng Li, Shigui Li, Zhiqi Lin, Junmei Yang, John Paisley, Delu Zeng

ICLR 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization

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
use learned minimum-path-variance schedule (MVP affine) instead of Linear pathdensity ratio estimation · MI estimation d=160 (MI=40)MSE: 9.17 -> 1.02Improved
use MVP (spherical) instead of VP pathdensity estimation · BSDS300test NLL: -131.90 -> -143.97Improved
use MVP (affine) instead of VP pathdensity estimation · POWERtest NLL: -0.04 -> -0.81Improved
use MVP (affine) instead of VP pathdensity estimation · GAStest NLL: -6.02 -> -6.95Improved
use MVP (spherical) instead of VP pathdensity estimation · MINIBOONEtest NLL: 18.25 -> 17.81Improved
use MVP (spherical) instead of VP pathdensity estimation · HEPMASStest NLL: 18.06 -> 18.01Improved
increase KMM components from 1 to 2 (ablation)MVP path · MI estimation d=160MSE: 1.32 -> 1.09Improved
increase KMM components from 1 to 2 (ablation)MVP path · BSDS300test NLL: -48.51 -> -145.44Improved
increase KMM components from 2 to 5 (ablation)MVP path · MI estimation d=160MSE: 1.09 -> 1.02Improved
increase KMM components from 5 to 8 (ablation)MVP path · MI estimation d=160MSE: 1.02 -> 2.32Hurt
increase KMM components from 5 to 8 (ablation)MVP path · BSDS300test NLL: -143.97 -> -143.60No clear effect
choose spherical instead of affine constraint (ablation)MVP path · BSDS300test NLL: -129.73 -> -143.97Improved

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