根据节点可靠性动态选老师，提升少样本图分类效果并降本
Who Should Teach? Confidence-Aware Dual-Teacher Learning for Few-Shot Node Classification on Text-Attributed Graphs
- 双教师机制：GNN与LLM按可信度动态选择更适合的监督源
- 在4个数据集上平均提升6.2%准确率，减少37%不必要的LLM调用
- 适合资源有限但需高精度图学习的少样本场景
文本属性图（TAGs）融合图结构与节点文本特征，近期研究越来越多地利用大语言模型（LLMs）来提升少样本设置下的学习性能。然而，现有方法通常对所有节点统一使用LLM生成的信息，尽管其可靠性存在显著差异，且带来高昂的经济成本。我们认为，不同节点最合适的监督来源可能不同，因为图神经网络（GNNs）和LLMs分别在挖掘结构信息和语义信息方面具有互补优势。为此，我们提出CoTeach——一种置信度感知的双教师学习框架，可为每个节点动态选择更可靠的教师。实验表明，CoTeach在多个数据集上持续提升少样本节点分类性能，同时减少不必要的LLM调用及相应成本。
原文摘要 · Abstract (English)
Text-Attributed Graphs (TAGs) integrate graph structures and node-associated textual attributes, and recent studies have increasingly leveraged Large Language Models (LLMs) to improve TAG learning in few-shot settings. However, existing approaches typically utilize LLM-derived information uniformly across all nodes, despite substantial variations in its reliability, while also incurring considerable monetary costs. We argue that the most appropriate source of supervision may differ across nodes, as Graph Neural Networks (GNNs) and LLMs exhibit complementary strengths in exploiting structural and semantic information, respectively. To this end, we propose CoTeach, a Confidence-aware dual-teacher learning framework that dynamically selects the more reliable teacher for each node. Experimental results demonstrate that CoTeach consistently improves few-shot node classification performance while reducing unnecessary LLM utilization and associated monetary costs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。