The paper "Efficient Graph Similarity Computation with Alignment Regularization" by Wei Zhuo and Guang Tan from the Sun Yat-sen University Shenzhen Campus proposes a novel approach to graph similarity computation (GSC) using Graph Neural Networks (GNNs). The authors address the computational challenges and high costs associated with traditional GSC methods, particularly those that rely on node-level matching modules. They introduce a regularization technique called Alignment Regularization (AReg), which simplifies the GSC process by eliminating the need for expensive node-to-node matching in the inference stage. AReg imposes a node-graph correspondence constraint on the GNN encoder during training, allowing the model to capture fine-grained interactions between graphs more efficiently. The paper also introduces a multi-scale Graph Edit Distance (GED) discriminator to enhance the discriminative ability of the learned representations. Extensive experiments on real-world datasets demonstrate the effectiveness, efficiency, and transferability of the proposed approach, showing superior performance compared to state-of-the-art methods. The framework, named ERIC (Efficient gRaph IMilarity Computation), is designed to be flexible and can be integrated into existing GSC models with minimal changes.The paper "Efficient Graph Similarity Computation with Alignment Regularization" by Wei Zhuo and Guang Tan from the Sun Yat-sen University Shenzhen Campus proposes a novel approach to graph similarity computation (GSC) using Graph Neural Networks (GNNs). The authors address the computational challenges and high costs associated with traditional GSC methods, particularly those that rely on node-level matching modules. They introduce a regularization technique called Alignment Regularization (AReg), which simplifies the GSC process by eliminating the need for expensive node-to-node matching in the inference stage. AReg imposes a node-graph correspondence constraint on the GNN encoder during training, allowing the model to capture fine-grained interactions between graphs more efficiently. The paper also introduces a multi-scale Graph Edit Distance (GED) discriminator to enhance the discriminative ability of the learned representations. Extensive experiments on real-world datasets demonstrate the effectiveness, efficiency, and transferability of the proposed approach, showing superior performance compared to state-of-the-art methods. The framework, named ERIC (Efficient gRaph IMilarity Computation), is designed to be flexible and can be integrated into existing GSC models with minimal changes.