GLIP-OOD: Zero-Shot Graph OOD Detection with Foundation Model

2025-04-29

AI Summary

This paper addresses the challenging problem of zero-shot node-level out-of-distribution (OOD) detection on graph-structured data—where no ID/OOD labels, node-level supervision, or prior knowledge of OOD class names is available. We propose the first zero-shot OOD detection framework synergizing Graph Foundation Models (GFMs) and Large Language Models (LLMs). Our method introduces: (1) LLM-driven semantic generation of transferable pseudo-OOD labels, overcoming dual bottlenecks of scarce pre-trained graph models and annotation dependency; and (2) a zero-shot vision–language alignment inference mechanism enabling fine-grained ID/OOD discrimination. Evaluated on four text-attributed graph benchmarks, our approach achieves state-of-the-art performance under fully unsupervised, label-free, and ID-training-free conditions—surpassing fully supervised SOTA methods and establishing the first genuine zero-shot node-level OOD detection for graph-structured data.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Out-of-distribution (OOD) detection is critical for ensuring the safety and reliability of machine learning systems, particularly in dynamic and open-world environments. In the vision and text domains, zero-shot OOD detection - which requires no training on in-distribution (ID) data - has made significant progress through the use of large-scale pretrained models such as vision-language models (VLMs) and large language models (LLMs). However, zero-shot OOD detection in graph-structured data remains largely unexplored, primarily due to the challenges posed by complex relational structures and the absence of powerful, large-scale pretrained models for graphs. In this work, we take the first step toward enabling zero-shot graph OOD detection by leveraging a graph foundation model (GFM). We show that, when provided only with class label names, the GFM can perform OOD detection without any node-level supervision - outperforming existing supervised methods across multiple datasets. To address the more practical setting where OOD label names are unavailable, we introduce GLIP-OOD, a novel framework that employs LLMs to generate semantically informative pseudo-OOD labels from unlabeled data. These labels enable the GFM to capture nuanced semantic boundaries between ID and OOD classes and perform fine-grained OOD detection - without requiring any labeled nodes. Our approach is the first to enable node-level graph OOD detection in a fully zero-shot setting, and achieves state-of-the-art performance on four benchmark text-attributed graph datasets.

Problem

• Enabling zero-shot OOD detection in graph-structured data
• Leveraging graph foundation models without node-level supervision
• Generating pseudo-OOD labels using LLMs for fine-grained detection

Innovation

• Leverages graph foundation model for zero-shot OOD detection
• Uses LLMs to generate pseudo-OOD labels from unlabeled data
• Achieves state-of-the-art performance without labeled nodes