AI Paper Atlas

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

Testing For Distribution Shifts with Conditional Conformal Test Martingales

Shalev Shaer, Yarin Bar, Drew Prinster, Yaniv Romano

ICML 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 0 citations (OpenAlex)

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 fixed-reference conditional CTM instead of growing-reference standard CTMconformal test martingale shift detector · Synthetic immediate shift N(0,1) to N(d,1), n=1000empirical power / detection delayImproved
use conditional CTM instead of standard CTM when shift occurs lateconformal test martingale shift detector · Synthetic delayed change-point (t0 = 4000)early detection powerMixed
use conditional CTM instead of standard CTMconformal test martingale shift detector · Synthetic gradual drift N(lambda*t, 1)empirical powerImproved
use conditional CTM instead of pairwise bettingGaussian AR(1) shift detection · Synthetic AR(1), a=0.7detection delayImproved
use conditional CTM instead of standard CTM with reference size n >= 500conformal test martingale shift detector · ImageNet-C (ViT-Base entropy score, all 15 corruptions)empirical powerImproved
use conditional CTM instead of standard CTM with reference size n = 100conformal test martingale shift detector · ImageNet-Cempirical powerHurt
add DKW-based estimation-error correction to the betting function (ablation)fixed-reference CTM · Type-I error vs reference sizetype-I error rateImproved
increase reference set size n (ablation)conditional CTM · ImageNet-Crejection-time ratioImproved
clip small betting parameter to zero (ablation)conditional CTM and standard CTM · Appendix F ablation on clipping CpowerImproved

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