Author name cluster

Fengjun Zhang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

AAAI Conference 2026 Conference Paper

Binary Message Passing for Generalizable Semi-Supervised Graph Anomaly Detection

Jingyuan Zhang
Xin Wang
Lei Yu
Li Yang
Fengjun Zhang

Graph Neural Networks (GNNs) have achieved impressive performance in semi-supervised graph anomaly detection (GAD). While many GNN variants have been developed for this task, they largely focus on advanced message aggregation schemes, leaving the message routing aspect underexplored. We argue that the commonly used broadcast-based routing can also hinder generalization, particularly in the presence of rare and structurally challenging (vertices with a high-degree) anomalies. To address this, we propose Binary Message Passing (BMP), a novel routing paradigm that models the message flow of each vertex as a binary tree (BMP tree), where vanilla graph convolution is decoupled by its left and right subtrees. Each vertex recursively gathers information from neighbors with higher anomaly probabilities within each subtree, thereby amplifying the propagation of anomaly information across the topology. The anomaly probabilities are estimated and updated by the model itself, enabling adaptive, self-supervised routing over iterations. Furthermore, combining multiple BMP trees into a BMP forest provides multi-scale structural context, enhancing the expressiveness of final vertex embeddings. Extensive experiments show that BMP improves detection performance under limited supervision while exhibiting better generalization across structurally diverse anomalies.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Restricted Global-Aware Graph Filters Bridging GNNs and Transformer for Node Classification

Jingyuan Zhang
Xin Wang
Lei Yu
Zhirong Huang
Li Yang
Fengjun Zhang

Transformers have been widely regarded as a promising direction for breaking through the performance bottlenecks of Graph Neural Networks (GNNs), primarily due to their global receptive fields. However, a recent empirical study suggests that tuned classical GNNs can match or even outperform state-of-the-art Graph Transformers (GTs) on standard node classification benchmarks. Motivated by this fact, we deconstruct several representative GTs to examine how global attention components influence node representations. We find that the global attention module does not provide significant performance gains and may even exacerbate test error oscillations. Consequently, we consider that the Transformer is barely able to learn connectivity patterns that meaningfully complement the original graph topology. Interestingly, we further observe that mitigating such oscillations enables the Transformer to improve generalization in GNNs. In a nutshell, we reinterpret the Transformer through the lens of graph spectrum and reformulate it as a global-aware graph filter with band-pass characteristics and linear complexity. This unique perspective introduces multi-channel filtering constraints that effectively suppress test error oscillations. Extensive experiments (17 homophilous, heterophilous graphs) provide comprehensive empirical evidence for our perspective. This work clarifies the role of Transformers in GNNs and suggests that advancing modern GNN research may still require a return to the graph itself.

PDF Details

ICML Conference 2025 Conference Paper

Towards Practical Defect-Focused Automated Code Review

Junyi Lu
Lili Jiang
Xiaojia Li
Jianbing Fang
Fengjun Zhang
Li Yang 0015
Chun Zuo

The complexity of code reviews has driven efforts to automate review comments, but prior approaches oversimplify this task by treating it as snippet-level code-to-text generation and relying on text similarity metrics like BLEU for evaluation. These methods overlook repository context, real-world merge request evaluation, and defect detection, limiting their practicality. To address these issues, we explore the full automation pipeline within the online recommendation service of a company with nearly 400 million daily active users, analyzing industry-grade C++ codebases comprising hundreds of thousands of lines of code. We identify four key challenges: 1) capturing relevant context, 2) improving key bug inclusion (KBI), 3) reducing false alarm rates (FAR), and 4) integrating human workflows. To tackle these, we propose 1) code slicing algorithms for context extraction, 2) a multi-role LLM framework for KBI, 3) a filtering mechanism for FAR reduction, and 4) a novel prompt design for better human interaction. Our approach, validated on real-world merge requests from historical fault reports, achieves a 2$\times$ improvement over standard LLMs and a 10$\times$ gain over previous baselines. While the presented results focus on C++, the underlying framework design leverages language-agnostic principles (e. g. , AST-based analysis), suggesting potential for broader applicability.

Details

EAAI Journal 2024 Journal Article

Perturbation-augmented Graph Convolutional Networks: A Graph Contrastive Learning architecture for effective node classification tasks

Qihang Guo
Xibei Yang
Fengjun Zhang
Taihua Xu

In the context of recent advances in Graph Convolutional Networks (GCNs) for semi-supervised learning, a significant highlight is the potential of Graph Contrastive Learning (GCL). Many established GCL-based GCNs focus on developing various augmentation strategies and learning architectures to efficiently generate and learn self-supervised information. Unfortunately, the node representation fusion is often neglected in the process of learning representation. To address the issue, we propose a novel end-to-end GCL framework called Perturbation-augmented Graph Convolutional Networks (PA-GCN), by simultaneously considering supervised information and self-supervised information of graphs to fuse node representations from the input graph and the augmentation graphs. Essentially, PA-GCN develops a perturbation augmentation strategy based on original graph relations, serving as a basis to generate diverse node information and self-supervised information. Additionally, PA-GCN introduces three main components: A encoder for node representation learning; an attention mechanism for effective node representation fusion; and a discriminator capable of learning self-supervised information from unlabeled nodes, thereby enabling joint learning of supervised information, node representation fusion, and self-supervised information. We conduct experiments on five real world datasets to evaluate the performance of PA-GCN in semi-supervised node classification tasks. The experimental results demonstrate the strong adaptability of PA-GCN for base encoders and the superiority of PA-GCN over the most advanced methods.

Details DOI