A novel data-level prompt injection attack against graph prompt learning
Open Access
- 20 September 2026
- journal article
- research article
- Published by Springer Nature in Cybersecurity
- Vol. 9 (1) , 220
Abstract
Graph Neural Networks (GNNs) are increasingly deployed in security-critical applications, including fraud detection, recommendation integrity analysis, and scientific knowledge mining. Recently, Graph Prompt Learning (GPL) has emerged as a parameter-efficient paradigm for adapting pretrained GNNs to downstream tasks. By optimizing only a small set of prompt parameters while keeping the pretrained encoder frozen, GPL enables efficient task adaptation under limited supervision. Despite its empirical success, the security implications of this paradigm remain largely unexplored. In this work, we investigate the vulnerability of GPL to data-level manipulations and introduce a novel threat, termed the data-level prompt injection attack. Unlike conventional attacks that rely on poisoning pretraining data or modifying model parameters, the proposed attack operates entirely at the downstream prompt learning stage. Specifically, an attacker injects a small number of carefully crafted malicious samples into the training dataset, without altering the pretrained encoder or the prompt learning algorithm. To realize this threat, we propose Graph Prompt Injection Attack (GPIA), which embeds carefully designed subgraph into training graphs to influence prompt optimization. Extensive experiments on multiple benchmarks show that GPIA achieves high attack success rates under low injection ratios while maintaining clean accuracy.Keywords
Funding Information
- Key Research and Development Plan Special Project of Henan Province Grant (241111211400)
- Luoyang Public Welfare Industry Scientific Research Plan Project (2602001A)
- National Natural Science Foundation of China Grant (61972133)
- Project of Leading Talents in Science and Technology Innovation for Thousands of People Plan in Henan Province Grant (204200510021)
This publication has 46 references indexed in Scilit:
- SIGFinger: A Subtle and Interactive GNN Fingerprinting Scheme Via Spatial Structure Inference PerturbationIEEE Transactions on Dependable and Secure Computing, 2025
- PSP: Pre-training and Structure Prompt Tuning for Graph Neural Networks2024
- Generalized Graph Prompt: Toward a Unification of Pre-Training and Downstream Tasks on GraphsIEEE Transactions on Knowledge and Data Engineering, 2024
- PoisonPrompt: Backdoor Attack on Prompt-Based Large Language Models2024
- GPL-GNN: Graph prompt learning for graph neural networkKnowledge-Based Systems, 2024
- G-Prompt: Graphon-based Prompt Tuning for graph classificationInformation Processing & Management, 2024
- A Causality-Aligned Structure Rationalization Scheme Against Adversarial Biased Perturbations for Graph Neural NetworksIEEE Transactions on Information Forensics and Security, 2023
- Exploratory Adversarial Attacks on Graph Neural Networks for Semi-Supervised Node ClassificationPattern Recognition, 2022
- Adversarial Attacks on Neural Networks for Graph Data2018
- Pitfalls2011