김정현/정소빈/강윤석’s paper has been accepted in

Title: PRISM:Reliable LLM Knowledge Distillation from Graph-Context Perspectives for Text-Attributed Graph Learning,
Author: Junghyun Kim, Sobin Jung, David Yoon Suk Kang and Sang-Wook Kim
Abstract
Text-attributed graph (TAG) learning aims to leverage both textual attributes and graph structures for node classification. Recent studies have utilized Large Language Models (LLMs) to generate pseudo-labels and rationales for few-label TAG learning. However, existing approaches typically rely solely on the textual attribute of target nodes, making the generated knowledge susceptible to noise and unreliability. To address this limitation, we propose PRISM(Perspective-based Reliable Imitation with Semantic Multiplicity), a reliability-aware knowledge distillation framework for few-label TAG learning. PRISM constructs multiple structural-context perspectives for each target node and estimates the reliability of LLMgenerated knowledge through cross-perspective consistency. The estimated reliability is then incorporated into the distillation process, enabling GNNs to selectively imitate more trustworthy supervision. Extensive experiments on five TAG benchmark datasets demonstrate that PRISM consistently outperforms existing methods in few-label node classification.