Saved in:
Bibliographic Details
Main Authors: Wen, Yanzhe, Li, Xunkai, Zhang, Qi, Lei, Zhu, Zeng, Guang, Li, Rong-Hua, Wang, Guoren
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.13989
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912384938934272
author Wen, Yanzhe
Li, Xunkai
Zhang, Qi
Lei, Zhu
Zeng, Guang
Li, Rong-Hua
Wang, Guoren
author_facet Wen, Yanzhe
Li, Xunkai
Zhang, Qi
Lei, Zhu
Zeng, Guang
Li, Rong-Hua
Wang, Guoren
contents Recently, large language models (LLMs) have significantly advanced text-attributed graph (TAG) learning. However, existing methods inadequately handle data uncertainty in open-world scenarios, especially concerning limited labeling and unknown-class nodes. Prior solutions typically rely on isolated semantic or structural approaches for unknown-class rejection, lacking effective annotation pipelines. To address these limitations, we propose Open-world Graph Assistant (OGA), an LLM-based framework that combines adaptive label traceability, which integrates semantics and topology for unknown-class rejection, and a graph label annotator to enable model updates using newly annotated nodes. Comprehensive experiments demonstrate OGA's effectiveness and practicality.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13989
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When LLMs meet open-world graph learning: a new perspective for unlabeled data uncertainty
Wen, Yanzhe
Li, Xunkai
Zhang, Qi
Lei, Zhu
Zeng, Guang
Li, Rong-Hua
Wang, Guoren
Machine Learning
Artificial Intelligence
Recently, large language models (LLMs) have significantly advanced text-attributed graph (TAG) learning. However, existing methods inadequately handle data uncertainty in open-world scenarios, especially concerning limited labeling and unknown-class nodes. Prior solutions typically rely on isolated semantic or structural approaches for unknown-class rejection, lacking effective annotation pipelines. To address these limitations, we propose Open-world Graph Assistant (OGA), an LLM-based framework that combines adaptive label traceability, which integrates semantics and topology for unknown-class rejection, and a graph label annotator to enable model updates using newly annotated nodes. Comprehensive experiments demonstrate OGA's effectiveness and practicality.
title When LLMs meet open-world graph learning: a new perspective for unlabeled data uncertainty
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.13989