AI Paper Atlas

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

Slimmed Asymmetrical Contrastive Learning and Cross Distillation for Lightweight Model Training

Jian Meng, Li Yang, Kyungmin Lee, Jinwoo Shin, Deliang Fan, Jae-sun Seo

NeurIPS 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Inductive bias and minimal representations · 1 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 cross-distillation (XD) to Barlow-Twins-style loss on 1x model, no teacherMobileNet-V3 · ImageNet-1K linear evaltop-1 acc (%): 56.70 -> 59.42Improved
Add asymmetrical slimmed contrastive learning (SACL) on top of XD (ablation)EfficientNet-B0 · ImageNet-1K linear evaltop-1 acc (%): 65.32Improved
Increase asymmetry from 2x-1x to 6x-1x (ablation)SACL+XD ResNet-20 · CIFAR-10 linear evalacc (%): 82.31 -> 86.81Improved
Train 100 vs 200 epochs with XD only (ablation)MobileNet-V3 · ImageNet-1Ktop-1 acc (%): 57.16 -> 59.42Improved
Extend training from 100 to 300 epochs (ablation)MobileNet-V1 XD · ImageNet-1Ktop-1 acc (%): 58.25Improved
Extend training to 300 epochs (ablation)MobileNet-V1 SACL+XD · ImageNet-1Ktop-1 acc (%): 60.63Improved
Replace inner-decorrelation distillation loss with negative-log loss (ablation)XD MobileNet-V1 · ImageNet-100acc (%): 80.30 -> 79.63Hurt
Use XD instead of Barlow TwinsMobileNet-V1 · ImageNet-100acc (%): 78.40 -> 80.30Improved
Use XD instead of Barlow TwinsResNet-50 · ImageNet-1K 300 epochstop-1 acc (%): 70.7 -> 71.1Improved
Use XD instead of DINOViT-Tiny · ImageNet-100acc (%): 63.04 -> 63.92Improved
Vary alpha weight of inner decorrelation loss (ablation)XD MobileNet-V1 · ImageNet-1Ktop-1 acc (%): best at alpha=0.9; within ± 1%No clear effect
Minimize only the infinitesimal inner correlation as LCD (ablation)SACL · trainingstability/accuracyHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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