AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Graph Contrastive Learning with Balanced Hard Negatives and Fine-grained Semantic-aware Positives

Hongshan Pu, Haoxu Zhang, Ye Liu, Hongmin Cai

AAAI 2026 · Graph learning › Graph self-supervised learning · approach: Graph contrastive learning analysis · 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 the BHN balanced hard negative module (ablation)BalanceGCL · ENZYMESaccuracy: 56.79 -> 60.29Improved
add the BHN balanced hard negative module (ablation)BalanceGCL · Synthieaccuracy: 74.35 -> 78.75Improved
add class-balance loss Lc instead of CGC KL loss (ablation)BalanceGCL · ENZYMESaccuracy: 57.34 -> 60.29Improved
add class-balance loss Lc instead of CGC KL loss (ablation)BalanceGCL · Synthieaccuracy: 76.20 -> 78.75Improved
add the FSP fine-grained positive module (ablation)BalanceGCL · ENZYMESaccuracy: 56.65 -> 60.29Improved
add the FSP fine-grained positive module (ablation)BalanceGCL · COLORS-3accuracy: 75.40 -> 80.18Improved
add positively contributing region perturbation (ablation)BalanceGCL · COLORS-3accuracy: 77.31 -> 80.18Improved
add negatively contributing region suppression (ablation)BalanceGCL · Synthieaccuracy: 77.50 -> 78.75Improved
add unimportant region random perturbation (ablation)BalanceGCL · ENZYMESaccuracy: 59.32 -> 60.29Improved
use both attribute and edge perturbation instead of attribute only (ablation)BalanceGCL-attr · ENZYMESaccuracy: 59.68 -> 60.29Improved
use both attribute and edge perturbation instead of edge only (ablation)BalanceGCL-edge · ENZYMESaccuracy: 58.37 -> 60.29Improved
use BalanceGCL instead of DRGCLDRGCL · COLORS-3accuracy: 78.90 -> 80.18Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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