AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

THGB: A Comprehensive Benchmark for Text-attributed Heterogeneous Graphs

Lixin Zhou, Zemin Liu, Yuan Fang, Dan Niu, Jing Ying

AAAI 2026 · Graph learning › Graph self-supervised learning · approach: LLMs and graph foundation models · 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
use HGNN (SimpleHGN) instead of homogeneous GCNGNN baselines · IMDB-TS few-shotMicro-F1: 0.6390 -> 0.6877Improved
use HGNN (SimpleHGN) instead of homogeneous GCNGNN baselines · Patent-TA few-shotMicro-F1: 0.5757 -> 0.6847Improved
use PLM-based THLM instead of best GNNnode classification · DBLP-TA few-shotMicro-F1: 0.7989 -> 0.8416Improved
use LLM-based HiGPT instead of PLM-based THLMnode classification · IMDB-TS few-shotMicro-F1: 0.7092 -> 0.7843Improved
use LLM-based OFA instead of THLMnode classification · DBLP-TA few-shotMicro-F1: 0.8416 -> 0.8823Improved
use heterogeneous models HGT/RGCN instead of simple GCN/GATlink prediction · IMDB-TSROC-AUC: GCN 0.7763 vs HGT 0.5373Hurt
use PLM/LLM-based methods instead of GNNslink prediction · ACM-TAROC-AUC: 0.7190 -> 0.9696Improved
remove features of non-target node types (use only target node text) (ablation)TAPE · IMDB-TS few-shotaccuracy / F1Hurt
use only titles or only storylines instead of both (ablation)feature variants · IMDB-TS few-shotaccuracy / F1Hurt
add text features of other node types to target text (ablation)feature variants · IMDB-TS few-shotaccuracy / F1Improved

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