AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

When to Retrain after Drift: A Data-Only Test of Post-Drift Data Size Sufficiency

Ren Fujiwara, Yasuko Matsubara, Yasushi Sakurai

ICLR 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 2 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 CALIPER-chosen window instead of incremental updatingMLP · MoCap, ADWINMSE: 412.606 -> 7.106Improved
use CALIPER-chosen window instead of incremental updatingMLP · Dysts, ADWINMSE: 71.75 -> 0.432Improved
use CALIPER instead of incremental updatingTransformer · Dysts, KSWINMSE: 0.582 -> 0.501Improved
use CALIPER instead of incremental updatingTransformer · Automobile, ADWINMSE: 1.866 -> 1.948Hurt
use CALIPER instead of incremental updatingTransformer · TEP, ADWINMSE: 2.376 -> 1.818Improved
use CALIPER instead of incremental updatingKRR · TEP, ADWINMSE: 1.780 -> 2.055Hurt
use CALIPER instead of fixed window 512MLP · MoCap, ADWINMSE: 23.887 -> 7.106Improved
use CALIPER instead of best fixed window 128 (KRR)KRR · TEP, ADWINMSE: 1.845 -> 2.055Hurt
use CALIPER instead of fixed window 512 (Automobile KRR)KRR · Automobile, ADWINMSE: 1.902 -> 1.949Hurt
vary ESS multiplier C in {2,3,4} (ablation)CALIPER MLP · all four datasetstest MAENo clear effect
increase theta_max to 18-20 (ablation)CALIPER MLP · MoCaptest MAEHurt
vary theta_max in {12..20} (ablation)CALIPER MLP · TEP, Automobile, Dyststest MAENo clear effect

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