AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

GFD-EMVC: Evolutionary Multi-View Classification with Label Noise via Gradient and Feature Dual-Perception

Shuai Li, Xinyan Liang, Yuhua Qian, Li Lv

ICML 2026 · Label-efficient and robust learning › Noisy labels · approach: Instance-dependent noise modeling

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
calibrate detected noisy labels instead of removing them (ablation)outlier gradient analysis (OGA) · CIFAR-10N Worsttest accuracy (%): 83.72 -> 84.46Improved
calibrate detected noisy labels instead of removing them (ablation)OGA · CIFAR-100N Noisy100test accuracy (%): 60.99 -> 61.47Improved
calibrate detected noisy labels instead of removing them (ablation)OGA · CIFAR-10N Aggregatetest accuracy (%): 91.36 -> 92.08Improved
train on purified dataEFB-EMVC · MVoxCeleb (40% label noise)test accuracy (%): 75.06 -> 80.55Improved
train on purified dataETMC · YoutubeFace (40% label noise)test accuracy (%): 52.23 -> 70.96Improved
train on purified dataMultiplication fusion · NUS (40% label noise)test accuracy (%): 26.76 -> 62.53Improved
train on purified dataETF · Reuters3 (40% label noise)test accuracy (%): 80.94 -> 81.31Improved
purify training data before evolutionary searchEMVC on YoutubeFace · fitness ranking of 15 individualsrank correlation (PC): 0.3857 (original vs purified ordering)Improved
increase clipping ratio kappa from 2.5% to 5% (ablation)GFD-EMVC · CIFAR-10N Aggregatetest accuracy (%): 91.64 -> 92.36Improved
increase clipping ratio kappa beyond noise rate (7.5% to 12.5%) (ablation)GFD-EMVC · CIFAR-10N Aggregatetest accuracy (%): 92.38 -> 92.17Hurt

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