arXiv 2409.011452024-09-02cs.SIcs.AI

LATEX-GCL：基于大语言模型（LLMs）的数据增强用于文本属性图对比学习

LATEX-GCL: Large Language Models (LLMs)-Based Data Augmentation for Text-Attributed Graph Contrastive Learning

University of Technology Sydney（悉尼科技大学）；The Hong Kong Polytechnic University（香港理工大学）；City University of Hong Kong（香港城市大学）；The Education University of Hong Kong（香港教育大学）

Haoran Yang, Xiangyu Zhao, Sirui Huang, Qing Li, Guandong Xu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出LATEX-GCL框架，利用大语言模型生成文本增强并发挥其自然语言处理能力，解决文本属性图对比学习中的信息损失、语义损失和增强不可控问题，在四个TAG数据集上验证了其优越性。

摘要：

图对比学习（GCL）是一种强大的自监督图学习范式，已在各种应用场景中引起广泛关注。然而，针对文本属性图（TAGs）上的GCL学习尚未得到探索，因为传统的增强技术（如特征嵌入掩码）无法直接处理TAGs上的文本属性。将GCL应用于TAGs的一种朴素策略是先通过语言模型将文本属性编码为特征嵌入，然后将嵌入输入后续GCL模块进行处理。这种策略面临三个关键挑战：I）无法避免信息损失，II）文本编码阶段存在语义损失，III）隐式增强约束导致结果不可控且难以理解。在本文中，我们提出了一种名为LATEX-GCL的新型GCL框架，利用大语言模型（LLMs）生成文本增强，并借助LLMs强大的自然语言处理（NLP）能力来解决上述三个局限性，从而为将GCL应用于TAG任务铺平道路。在四个高质量TAG数据集上的大量实验表明了所提出的LATEX-GCL方法的优越性。源代码和数据集已发布以便于复现，可通过以下链接访问：https://anonymous.4open.science/r/LATEX-GCL-0712。

英文摘要：

Graph Contrastive Learning (GCL) is a potent paradigm for self-supervised graph learning that has attracted attention across various application scenarios. However, GCL for learning on Text-Attributed Graphs (TAGs) has yet to be explored. Because conventional augmentation techniques like feature embedding masking cannot directly process textual attributes on TAGs. A naive strategy for applying GCL to TAGs is to encode the textual attributes into feature embeddings via a language model and then feed the embeddings into the following GCL module for processing. Such a strategy faces three key challenges: I) failure to avoid information loss, II) semantic loss during the text encoding phase, and III) implicit augmentation constraints that lead to uncontrollable and incomprehensible results. In this paper, we propose a novel GCL framework named LATEX-GCL to utilize Large Language Models (LLMs) to produce textual augmentations and LLMs' powerful natural language processing (NLP) abilities to address the three limitations aforementioned to pave the way for applying GCL to TAG tasks. Extensive experiments on four high-quality TAG datasets illustrate the superiority of the proposed LATEX-GCL method. The source codes and datasets are released to ease the reproducibility, which can be accessed via this link: https://anonymous.4open.science/r/LATEX-GCL-0712.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。