GLGaichao Li

Gaichao Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Gaichao Li has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Global-Local Contrastive Graph Transformer for Node Classification

    by Gaichao Li, Jinsong Chen, Kun He

    Graph Transformers have shown promising performance in node classification tasks. Rather than using the entire graph as input, recent approaches focus on constructing token sequences for target nodes, enhancing scalability and reducing noise in large graphs. However, these token sequence-based graph Transformers have three key limitations: (1) limited ability to capture global information while missing fine-grained insights from distant but relevant nodes; (2) insufficient consideration of the interplay between topology and attributes in local context sampling; (3) potential interference caused by mixing different token types within the same sequence. To address these challenges, we propose a novel global-local contrastive graph Transformer termed GLFormer, which learns node representations more comprehensively by integrating global and local graph information. From the global perspective, we leverage graph hierarchies to build layer-wise global sequences that capture coarse-to-fine global information. From the local perspective, we construct local sequences by sampling nodes from a weighted graph derived from attribute similarity, effectively preserving both topological and attributed information. Additionally, we introduce a global-local alignment loss, which aligns node representations learned from different token sequences, further aiding the expressiveness of target node representation. Empirical results demonstrate that GLFormer consistently outperforms mainstream baselines across 13 widely-used graph benchmarks.

  • Open Access

    Tokenized Heterogeneous Graph Transformer with Enhanced Local and Global Representation Learning

    by Gaichao Li, Jinsong Chen, Yangzhe Peng, Kun He

    Graph Transformers have demonstrated superiority in handling complex heterogeneous graphs in recent years. However, existing models still face several inherent challenges: (1) reliance on manually designed meta-paths to encode explicit local graph heterogeneity; (2) inability to capture fine-grained global information from distant yet relevant nodes. To address these limitations, we introduce THFormer, a novel node tokenized heterogeneous graph Transformer that learns expressive node representations by incorporating local and global perspectives. From the local perspective, we employ multiple subsequences for different heterogeneous types to explicitly encode local semantic relations, eliminating the need for manually designed meta-paths. From the global perspective, we design a local masking and global sampling mechanism to construct global structural (semantic) sequences, effectively capturing fine-grained global structural (semantic) information. Subsequently, THFormer separately feeds the resulting global and local sequences into standard Transformer layers as model inputs. Since these sequences represent two distinct views of the same target node, their corresponding outputs are naturally aligned to generate self-supervisory signals for model training, further enhancing the expressiveness and reliability of the target node representation. Extensive experiments are conducted to validate the efficacy of THFormer, and the quantitative performance gains are 0.24%, 0.31%, 0.51%, and 0.81% on DBLP, ACM, IMDB, and Freebase, respectively. The experimental results demonstrate the superiority of THFormer over representative heterogeneous graph neural networks and graph Transformer models.

  • GTPool: Graph Transformer Pooling With Diverse Sampling

    by Gaichao Li, Jinsong Chen, Kun He

    Graph pooling techniques have emerged as powerful tools for downsampling graphs, yielding impressive results on various graph-level tasks like graph classification and generation. Node dropping pooling stands out as a significant approach in this domain, utilizing learnable scoring functions to drop nodes with relatively lower significance. However, previous node dropping methods face two critical limitations: (1) they often struggle to capture long-range dependencies effectively for each pooled node, potentially limiting the expressiveness of node representations. (2) by exclusively retaining the highest-scoring nodes, they tend to preserve similar nodes, thereby discarding valuable information residing in low-scoring nodes; To address these issues, we propose a novel Graph Transformer Pooling method termed GTPool. GTPool constructs a node dropping pooling layer based on the Transformer architecture, aiming to efficiently capture long-range pairwise interactions while promoting diverse sampling. A key component of GTPool is its scoring module, grounded in the self-attention mechanism, which assesses the relevance between node representations and the graph representation. This nuanced approach offers a more natural means of measuring node importance. Additionally, GTPool adopts Roulette Wheel Sampling (RWS), a diversified sampling method that is capable of preserving nodes across various scoring intervals, rather than solely focusing on higher-scoring nodes. Consequently, GTPool can effectively capture long-range information and identify more representative nodes, thereby outperforming existing popular graph pooling methods across 14 benchmark datasets.


  • NAGphormer+: A Tokenized Graph Transformer With Neighborhood Augmentation for Node Classification in Large Graphs

    by Jinsong Chen, Chang Liu, Kaiyuan Gao, Gaichao Li, Kun He

    Graph Transformers, emerging as a new architecture for graph representation learning, suffer from the quadratic complexity and can only handle graphs with at most thousands of nodes. To this end, we propose a Neighborhood Aggregation Graph Transformer (NAGphormer) that treats each node as a sequence containing a series of tokens constructed by our proposed Hop2Token module. For each node, Hop2Token aggregates the neighborhood features from different hops into different representations, producing a sequence of token vectors as one input. In this way, NAGphormer could be trained in a mini-batch manner and thus could scale to large graphs with millions of nodes. To further enhance the model's generalization, we propose NAGphormer+, an extended model of NAGphormer with a novel data augmentation method called Neighborhood Augmentation (NrAug). Based on the output of Hop2Token, NrAug simultaneously augments the features of neighborhoods from global as well as local views. In this way, NAGphormer+ can fully utilize the neighborhood information of multiple nodes, thereby undergoing more comprehensive training and improving the model's generalization capability. Extensive experiments on benchmark datasets from small to large demonstrate the superiority of NAGphormer+ against existing graph Transformers and mainstream GNNs, as well as the original NAGphormer.