OpenRTAG: A Comprehensive Benchmark for Robust Text-Attributed Graph Learning under Data Quality Degradation

2026-07-21

AI Summary

This study addresses the significant performance degradation of models on real-world textual attribute graphs (TAGs), which often suffer from poor data quality—including text sparsity, structural noise, and label imbalance—while noting the absence of systematic robustness evaluation under multidimensional degradation scenarios. To bridge this gap, the work proposes a unified 3×3 degradation taxonomy encompassing nine representative and composite data-quality degradation settings, and establishes a standardized benchmark across nine TAG datasets and three downstream tasks. Through comprehensive evaluation of conventional GNNs, LLM-enhanced GNNs, and graph foundation models under varying degradation conditions, the study systematically analyzes their effectiveness, efficiency, and robustness, revealing distinct sensitivity patterns. These empirical insights offer practical guidance for model selection and improvement in low-quality real-world TAG applications, thereby filling a critical void in systematic robustness assessment for this domain.

Abstract

Text-attributed graphs (TAGs) are an important graph data form that combine relational structure with rich node text. However, real-world TAGs are often imperfect, with quality issues arising from text, structure, and labels, and typically manifesting as sparsity, noise, and imbalance. These dimensions define nine representative degradation scenarios that can substantially affect TAG learning. Although prior studies have explored specific mitigation strategies, existing evidence remains fragmented across degradation types, datasets, tasks, and model families, leaving TAG robustness insufficiently understood. To address this gap, we present OpenRTAG, a robustness benchmark for text-attributed graph learning. OpenRTAG organizes TAG quality issues into a unified 3 * 3 taxonomy and supports standardized evaluation across nine TAG datasets and three downstream tasks. It systematically evaluates scenario validity and model sensitivity, compares traditional GNNs, LLM-GNNs, and a representative GFM, investigates the effectiveness, efficiency, and robustness of scenario-matched baselines, and further examines model behavior under composite degradation scenarios. OpenRTAG provides a standardized testbed for understanding robustness in TAG learning under realistic low-quality settings.

Problem

• text-attributed graphs
• data quality degradation
• robustness benchmark
• graph learning
• low-quality data

Innovation

• Text-Attributed Graphs
• Robustness Benchmark
• Data Quality Degradation
• Graph Neural Networks
• LLM-GNN