AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Evaluating Sample Utility for Efficient Data Selection by Mimicking Model Weights

Tzu-Heng Huang, Manjot Bilkhu, John Cooper, Frederic Sala, Javier Movellan

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 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
re-weight batch samples by mimic scoremini-batch SGD · six noisy image datasets, noise 0.5 (avg)test accuracy (%): 69.46 -> 74.53Improved
weight samples by gradient norm (GraNd)mini-batch SGD · six noisy image datasets, noise 0.5 (avg)test accuracy (%): 69.46 -> 53.36Hurt
use mimic score instead of Rho-Loss excess lossRho-Loss · six noisy image datasets, noise 0.6 (avg)test accuracy (%): 71.93 -> 73.01Improved
use mimic score with hold-out reference modelRho-Loss · CIFAR100 50% noisetest accuracy (%): 75.11 -> 75.81Improved
use pretrained reference weights instead of a labeled validation set (ablation)Grad-Mimic · six datasets, 50% noisetest accuracyImproved
add mimic-score re-weighting to CLIP trainingCLIP from scratch · DataComp small38-task average: 0.131 -> 0.146Improved
add mimic-score re-weighting to CLIP trainingCLIP from scratch · DataComp medium38-task average: 0.254 -> 0.258Improved
use image-encoder last layer instead of text-encoder as reference (ablation)Grad-Mimic · DataComp small38-task average: 0.139 -> 0.145Improved
mimic-score training convergenceCLIP from scratch · DataComptraining steps: 20.7% fewer stepsImproved
remove mimic-score bottom 5% from CLIP Score top 30% filterCLIP Score filter · DataComp medium38-task average: 0.321 -> 0.323Improved
remove mimic-score bottom 20% from CLIP Score filterCLIP Score filter · DataComp small38-task average: 0.165 -> 0.164No clear effect
use mimic score instead of GraNd/AGRA/influence for costGrad-Mimic · per-batch overheadcompute: 2.6x lower overheadImproved

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