AI Paper Atlas

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

Front-Loaded Robust Conformal Prediction: Heavy Calibration, Minimal Test-Time Cost

Soroush H. Zargarbashi, Mohammad Sadegh Akhondzadeh, Aleksandar Bojchevski

ICML 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Conformal prediction sets

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
replace RCP1 with FLoR(1) (heavy calibration sampling)RCP1 · ImageNet ResNet r=0.50avg set size: 144.226 -> 24.076Improved
replace RCP1 with FLoR(1)RCP1 · ImageNet ResNet r=0.25avg set size: 45.019 -> 15.905Improved
replace RCP1 with FLoR(1)RCP1 · ImageNet Diffusion+ViT r=0.50avg set size: 197.633 -> 20.765Improved
replace RCP1 with FLoR(1)RCP1 · CIFAR-10 ResNet r=0.50avg set size: 5.413 -> 4.165Improved
replace RCP1 with FLoR(1)RCP1 · CIFAR-10 Diffusion+ViT r=0.50avg set size: 8.935 -> 6.453Improved
replace RCP1 with FLoR(1) at zero radiusRCP1 · ImageNet ResNet r=0.00avg set size: 8.605 -> 12.451Hurt
replace RCP1 with FLoR(1) at zero radiusRCP1 · CIFAR-10 ResNet r=0.00avg set size: 2.470 -> 2.680Hurt
disable finite-sample correction in FLoR(1) (APS score) (ablation)FLoR(1) · ImageNet/CIFAR APS, ResNetavg set sizeImproved
increase calibration-time MC sample rate (ablation)FLoR(1) · CIFAR-10 ResNet r=0.5avg set sizeImproved
calibrate FLoR(k) for majority vote at low test-time sample rate (21 to 101)BinCP · CIFAR-10 ResNetavg set sizeImproved
use plain majority vote of test-time sets without calibration awarenessRCP1/FLoR(1) · coverage guaranteecoverage: 1 - alpha -> 1 - 2 alphaHurt
increase smoothing sigma (ablation)FLoR(1) · ImageNet/CIFAR-10avg set sizeHurt

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