AI Paper Atlas

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

GenView: Enhancing View Quality with Pretrained Generative Model for Self-Supervised Learning

Xiaojie Li, Yibo Yang, Xiangtai Li, Jianlong Wu, Yue Yu, Bernard Ghanem, Min Zhang

ECCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Inductive bias and minimal representations · 4 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

add quality-driven contrastive loss to framework · IN-100 linear eval · top-1 accuracy: 71.50 -> 74.88 Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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 GenView to MoCov2MoCov2 ResNet-50 · ImageNet-1K linear evaltop-1 accuracy: 67.5 -> 70.0Improved
add GenView to MoCov2MoCov2 ResNet-50 · ImageNet-1K semi-supervised 1% labelstop-1 accuracy: 42.1 -> 50.6Improved
add GenView to BYOLBYOL ResNet-50 · MS-COCO detectionAP: 40.6 -> 41.2Improved
add 0.3M Laion400M images to ImageNet-1K pretraining (ablation)MoCov3 ResNet-50 · ImageNet-1K linear eval (50 ep)top-1 accuracy: 62.39 -> 63.31Improved
use GenView framework with 0.15M generated images instead of adding 0.3M extra real or synthetic images (ablation)MoCov3 ResNet-50 · ImageNet-1K linear eval (50 ep)top-1 accuracy: 62.39 -> 65.62Improved
add generated-view framework without adaptive noise or quality loss (ablation)MoCov3 ResNet-18 · IN-100 linear evaltop-1 accuracy: 65.52 -> 71.50Improved
add adaptive view generation to framework (ablation)GenView · IN-100 linear evaltop-1 accuracy: 71.50 -> 73.96Improved
add quality-driven contrastive loss to framework (ablation)GenView · IN-100 linear evaltop-1 accuracy: 71.50 -> 74.88Improved
add quality-driven contrastive loss to baseline without generated views (ablation)MoCov3 ResNet-18 · IN-100 linear evaltop-1 accuracy: 65.52 -> 66.97Improved
use adaptive noise level selection instead of constant or random (ablation)GenView · IN-100 linear evaltop-1 accuracy: 72.96 -> 73.96Improved
raise noise level constant from 0 to 100-400 (ablation)GenView · IN-100 linear evaltop-1 accuracy: 71.80 -> 72.14 (CS 100)No clear effect
increase probability of applying GenView (ablation)GenView · ImageNet-1K linear eval (50 ep)top-1 accuracy: 62.39 -> 70.55Improved

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