LLM-Enhanced Energy Contrastive Learning for Out-of-Distribution Detection in Text-Attributed Graphs
This addresses the challenge of maintaining performance in real-world networks like citation or social graphs when faced with OOD data, representing an incremental improvement by combining LLMs with contrastive learning for a specific task.
The paper tackles the problem of node-level out-of-distribution (OOD) detection in text-attributed graphs, where existing methods degrade with OOD data, and proposes LECT, which integrates large language models and energy-based contrastive learning to achieve high classification accuracy and robust OOD detection, outperforming state-of-the-art baselines on six benchmark datasets.
Text-attributed graphs, where nodes are enriched with textual attributes, have become a powerful tool for modeling real-world networks such as citation, social, and transaction networks. However, existing methods for learning from these graphs often assume that the distributions of training and testing data are consistent. This assumption leads to significant performance degradation when faced with out-of-distribution (OOD) data. In this paper, we address the challenge of node-level OOD detection in text-attributed graphs, with the goal of maintaining accurate node classification while simultaneously identifying OOD nodes. We propose a novel approach, LLM-Enhanced Energy Contrastive Learning for Out-of-Distribution Detection in Text-Attributed Graphs (LECT), which integrates large language models (LLMs) and energy-based contrastive learning. The proposed method involves generating high-quality OOD samples by leveraging the semantic understanding and contextual knowledge of LLMs to create dependency-aware pseudo-OOD nodes, and applying contrastive learning based on energy functions to distinguish between in-distribution (IND) and OOD nodes. The effectiveness of our method is demonstrated through extensive experiments on six benchmark datasets, where our method consistently outperforms state-of-the-art baselines, achieving both high classification accuracy and robust OOD detection capabilities.