AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Maximizing Incremental Information Entropy for Contrastive Learning

Jiansong Zhang, Zhuoqin Yang, Xu Wu, Xiaoling Luo, Peizhong Liu, Linlin Shen

ICLR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Inductive bias and minimal representations · 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
add SAIB entropy-generation module (ablation)MoCo-v2 ResNet-18 · ImageNet-1ktop-1 accuracy (%): 52.50 -> 58.80Improved
add KL regularization (ablation)MoCo-v2 + SAIB · ImageNet-1ktop-1 accuracy (%): 58.80 -> 59.15Improved
add spectral-norm encoder regularizer (ablation)MoCo-v2 + SAIB + KL · ImageNet-1ktop-1 accuracy (%): 59.15 -> 59.41Improved
cascade 2 or 3 SAIB modules instead of 1 (ablation)IE-CL · ImageNet-1ktop-1 accuracy (%): 59.41 -> 58.62 (2x), 58.71 (3x)Hurt
add SAIB to MoCo-v2 (ablation)MoCo-v2 · ImageNet-100top-1 accuracy (%): 66.29 -> 67.54Improved
add SAIB to BYOL (ablation)BYOL · ImageNet-100top-1 accuracy (%): 67.95 -> 68.76Improved
add SAIB to SimCLR (ablation)SimCLR · ImageNet-100top-1 accuracy (%): 63.34 -> 64.02Improved
add SAIB to SimSiam (ablation)SimSiam · ImageNet-100top-1 accuracy (%): 66.25 -> 66.97Improved
use IE-CL instead of Matrix-SSL at 100 epochsResNet-50 pretraining · ImageNet linear evaltop-1 accuracy (%): 69.2 -> 68.3Hurt
use IE-CL for 800 epochsResNet-50 pretraining · ImageNet linear evaltop-1 accuracy (%): 71.9 -> 73.2Improved

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