用大模型分离图数据中的信号与噪声，提升对比学习效果
Disentangle-then-Refine: LLM-Guided Decoupling and Structure-Aware Refinement for Graph Contrastive Learning
- 用大模型主动解析属性，分出有用信号和噪声
- 通过频谱特性筛选保留信号，消除大模型幻觉
- 适合处理带文本的图数据，尤其关注精度与效率
传统文本属性图上的图对比学习依赖盲目的随机增强，无意中将任务相关信号与噪声混合。我们提出SDM-SCR框架，基于近似正交分解。首先，语义解耦模块（SDM）利用大语言模型的指令遵循能力，主动将原始属性解析为不对称的任务导向信号视图和噪声视图，实现从随机扰动到语义感知解耦的范式转变。随后，语义一致性正则化（SCR）利用谱观察：语义信号在拓扑上平滑，而残余噪声为高频。SCR作为选择性谱滤波器，仅在信号子空间强制一致性，消除大模型幻觉且不造成过度平滑。这种‘解耦-精炼’机制确保了严格的信号净化。大量实验表明，SDM-SCR在准确率与效率上均达到当前最优水平。
原文摘要 · Abstract (English)
Conventional Graph Contrastive Learning (GCL) on Text-Attributed Graphs (TAGs) relies on blind stochastic augmentations, inadvertently entangling task-relevant signals with noise. We propose SDM-SCR, a robust framework anchored in Approximate Orthogonal Decomposition. First, the Semantic Decoupling Module (SDM) leverages the instruction-following capability of Large Language Models (LLMs) to actively parse raw attributes into asymmetric, task-oriented signal and noise views. This shifts the paradigm from random perturbation to semantic-aware disentanglement. Subsequently, Semantic Consistency Regularization (SCR) exploits the spectral observation that semantic signals are topologically smooth while residual noise is high-frequency. SCR functions as a selective spectral filter, enforcing consistency only on the signal subspace to eliminate LLM hallucinations without over-smoothing. This ``Disentangle-then-Refine'' mechanism ensures rigorous signal purification. Extensive experiments demonstrate that SDM-SCR achieves SOTA performance in accuracy and efficiency.
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。