4 papers
Revealing Modular Gradient Noise Imbalance in LLMs: Calibrating Adam via Signal-to-Noise Ratio
Ziqing Wen, Zhouyang Liu, Jiahuan Wang +4
The impressive performance of large language models (LLMs) arises from their massive scale and heterogeneous module composition. However, this structural heterogeneity introduces a…
Rethinking Flexible Graph Similarity Computation: One-step Alignment with Global Guidance
Zhouyang Liu, Ning Liu, Yixin Chen +3
Graph Edit Distance (GED) is a widely used measure of graph similarity, valued for its flexibility in encoding domain knowledge through operation costs. However, existing learning-…
Hierarchy-Aware Neural Subgraph Matching with Enhanced Similarity Measure
Zhouyang Liu, Ning Liu, Yixin Chen +3
Subgraph matching is challenging as it necessitates time-consuming combinatorial searches. Recent Graph Neural Network (GNN)-based approaches address this issue by employing GNN en…
Graph2Region: Efficient Graph Similarity Learning with Structure and Scale Restoration
Zhouyang Liu, Yixin Chen, Ning Liu +2
Graph similarity is critical in graph-related tasks such as graph retrieval, where metrics like maximum common subgraph (MCS) and graph edit distance (GED) are commonly used. Howev…