Large Language Models for Requirements Engineering: A Cross-Task Empirical Evaluation
研究通过两个实证研究评估了大型语言模型在需求工程中的应用，覆盖了从需求分类到可追溯性链接识别等五项活动，旨在解决需求信息提取的难题。
0 citationsRead paper
Selected work
Representative Papers
Opinion polarization from compression-based decision making where agents optimize local complexity and global simplicity
This study investigates how individual cognitive preferences drive social opinion polarization. By developing an agent-based model that, for the first time, integrates the mechanisms of optimal distinctiveness—seeking local diversity—and cognitive compression—simplifying global information—the authors quantify the trade-off between these two processes using Shannon entropy. The model reveals that opinion clusters evolve dynamically rather than remaining fixed during polarization, and that moderate local group sizes are most conducive to polarized outcomes. High levels of cognitive compression increase system unpredictability, whereas low compression yields stable group structures. The results successfully reproduce heterogeneous opinion clusters observed in real-world settings and link conditions for polarization to Dunbar’s number, offering a novel perspective on how group size influences the emergence of polarization.
0 citationsRead paper
MAGNET: A Multi-Graph Attentional Network for Code Clone Detection
Existing code clone detection methods typically rely on a single program representation—e.g., AST, CFG, or DFG—failing to comprehensively capture program semantics; hybrid approaches are limited by handcrafted fusion strategies and yield marginal improvements. This paper proposes a novel multi-graph attention framework that unifies AST, CFG, and DFG for fine-grained semantic representation. Its core innovations include: (1) a gated cross-attention mechanism enabling dynamic inter-graph interaction; (2) residual graph neural networks combined with node-level self-attention to jointly model local and long-range dependencies; and (3) Set2Set pooling to generate robust program-level embeddings. Evaluated on BigCloneBench and Google Code Jam, our method achieves F1 scores of 96.5% and 99.2%, respectively—substantially surpassing state-of-the-art baselines. Ablation studies confirm the effectiveness and synergistic contributions of each component.
0 citationsRead paper
AST-Enhanced or AST-Overloaded? The Surprising Impact of Hybrid Graph Representations on Code Clone Detection
Code cloning is a major contributor to high maintenance costs and security risks; however, existing AST-based deep learning approaches suffer from insufficient semantic representation. Method: This paper systematically evaluates the effectiveness of various graph representations—including ASTs, CFGs, DFGs, and FA-ASTs—and their combinations, in conjunction with GNN architectures (GCN, GAT, GMN) for code clone detection. Contribution/Results: We reveal a strong coupling between graph fusion strategies and model architecture: AST+CFG+DFG significantly improves accuracy for GCN and GAT, whereas FA-AST degrades performance due to structural redundancy; remarkably, GMN achieves superior performance using AST alone, outperforming most fused variants. Our approach attains 98.2% accuracy across multiple benchmarks. This work establishes, for the first time, the optimal matching relationships between GNN architectures and graph representations, providing a reusable, principled modeling guideline for industrial-scale code clone detection.
0 citationsRead paper
Recent publications
Latest Papers
Large Language Models for Requirements Engineering: A Cross-Task Empirical Evaluation
研究通过两个实证研究评估了大型语言模型在需求工程中的应用，覆盖了从需求分类到可追溯性链接识别等五项活动，旨在解决需求信息提取的难题。
0 citationsRead paper
Opinion polarization from compression-based decision making where agents optimize local complexity and global simplicity
This study investigates how individual cognitive preferences drive social opinion polarization. By developing an agent-based model that, for the first time, integrates the mechanisms of optimal distinctiveness—seeking local diversity—and cognitive compression—simplifying global information—the authors quantify the trade-off between these two processes using Shannon entropy. The model reveals that opinion clusters evolve dynamically rather than remaining fixed during polarization, and that moderate local group sizes are most conducive to polarized outcomes. High levels of cognitive compression increase system unpredictability, whereas low compression yields stable group structures. The results successfully reproduce heterogeneous opinion clusters observed in real-world settings and link conditions for polarization to Dunbar’s number, offering a novel perspective on how group size influences the emergence of polarization.
0 citationsRead paper
MAGNET: A Multi-Graph Attentional Network for Code Clone Detection
Existing code clone detection methods typically rely on a single program representation—e.g., AST, CFG, or DFG—failing to comprehensively capture program semantics; hybrid approaches are limited by handcrafted fusion strategies and yield marginal improvements. This paper proposes a novel multi-graph attention framework that unifies AST, CFG, and DFG for fine-grained semantic representation. Its core innovations include: (1) a gated cross-attention mechanism enabling dynamic inter-graph interaction; (2) residual graph neural networks combined with node-level self-attention to jointly model local and long-range dependencies; and (3) Set2Set pooling to generate robust program-level embeddings. Evaluated on BigCloneBench and Google Code Jam, our method achieves F1 scores of 96.5% and 99.2%, respectively—substantially surpassing state-of-the-art baselines. Ablation studies confirm the effectiveness and synergistic contributions of each component.
0 citationsRead paper
AST-Enhanced or AST-Overloaded? The Surprising Impact of Hybrid Graph Representations on Code Clone Detection
Code cloning is a major contributor to high maintenance costs and security risks; however, existing AST-based deep learning approaches suffer from insufficient semantic representation. Method: This paper systematically evaluates the effectiveness of various graph representations—including ASTs, CFGs, DFGs, and FA-ASTs—and their combinations, in conjunction with GNN architectures (GCN, GAT, GMN) for code clone detection. Contribution/Results: We reveal a strong coupling between graph fusion strategies and model architecture: AST+CFG+DFG significantly improves accuracy for GCN and GAT, whereas FA-AST degrades performance due to structural redundancy; remarkably, GMN achieves superior performance using AST alone, outperforming most fused variants. Our approach attains 98.2% accuracy across multiple benchmarks. This work establishes, for the first time, the optimal matching relationships between GNN architectures and graph representations, providing a reusable, principled modeling guideline for industrial-scale code clone detection.
0 citationsRead paper