AI Paper Atlas

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

Self-Supervised Contrastive Re-Learning for Multi-Graph Multi-Label Classification

Meixia Wang, Yuhai Zhao, Zhengkui Wang, Yejiang Wang, Miaomiao Huang, Fenglong Ma, Fazal Wahab, Wen Shan, Xingwei Wang

AAAI 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Contrastive versus non-contrastive theory · 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 graph-level contrastive loss L1 (ablation)PETAL · Dblp(ai)Average Precision: 0.778 -> 0.780Improved
add bag-level contrastive loss L2 (ablation)PETAL · Dblp(ai)Average Precision: 0.772 -> 0.780Improved
add cross-granularity bag-graph loss L3 (ablation)PETAL · Dblp(ai)Average Precision: 0.752 -> 0.780Improved
add contrastive re-learning to generate augmentation-free positives (ablation)PETAL · Dblp(ai)Average Precision: 0.430 -> 0.780Improved
add label prototypes C (ablation)PETAL · Dblp(ai)Average Precision: 0.721 -> 0.780Improved
add contrastive re-learning (ablation)PETAL · Dblp(ai)Macro-F1: 0.109 -> 0.386Improved
use PETAL instead of supervised cfMGNMLcfMGNML · Dblp(ai)Average Precision: 0.749 -> 0.780Improved
use PETAL instead of supervised cfMGNMLcfMGNML · MSRC v2Average Precision: 0.659 -> 0.696Improved
use PETAL instead of graph contrastive learning baseline PaGCLPaGCL · Dblp(ai)Average Precision: 0.455 -> 0.739Improved
use PETAL instead of PaGCLPaGCL · Dblp(db)Average Precision: 0.470 -> 0.713Improved
use PETAL instead of PaGCLPaGCL · VOC12One Error: 0.285 -> 0.278No clear effect

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