AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

A Comprehensive Study on Text-attributed Graphs: Benchmarking and Rethinking

Hao Yan, Chaozhuo Li, Ruosong Long, Chao Yan, Jianan Zhao, Wenwen Zhuang, Jun Yin, Peiyan Zhang, Weihao Han, Hao Sun, Weiwei Deng, Qi Zhang and 3 more

NeurIPS 2023 · Graph learning › Graph self-supervised learning · approach: LLMs and graph foundation models · 22 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 topological contrastive pre-training (TCL) to the PLM (ablation)BERT-Base · ogbn-arxiv-TAaccuracy: 72.96 -> 74.87Improved
add topological masked LM pre-training (TMLM) (ablation)BERT-Base · ogbn-arxiv-TAaccuracy: 72.96 -> 73.97Improved
add Deepwalk-alignment pre-training (TDK) (ablation)BERT-Base · ogbn-arxiv-TAaccuracy: 72.96 -> 74.23Improved
combine TMLM then TCL+TDK (TMDC) instead of single task (ablation)BERT-Base · ogbn-arxiv-TAaccuracy: 74.87 -> 76.11Improved
add TMDC pre-trainingBERT-Base · Photoaccuracy: 77.53 -> 84.09 (best GNN 84.46)Mixed
add TMDC pre-trainingBERT-Base · Childrenaccuracy: 59.91 -> 61.43Improved
co-train BERT-Tiny with GCN instead of static featuresGCN · ogbn-arxiv-TAaccuracy: 72.03 -> 69.22Hurt
co-train BERT-Tiny with GraphSAGE instead of static featuresGraphSAGE · Photoaccuracy: 83.25 -> 86.64Improved
co-train BERT-Tiny with GCN instead of static featuresGCN · Computersaccuracy: 87.43 -> 83.93Hurt
increase sampled neighbors per layer in co-training (ablation)GCN(T)/SAGE(T) · ogbn-arxiv-TA, History, Computers, PhotoaccuracyImproved
use DeBERTa instead of BERT/RoBERTa as static feature encoderRevGAT · ogbn-arxiv-TAaccuracy: 74.82 -> 68.26 (Base RoBERTa vs DeBERTa)Hurt

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