Debugging Tabular Logs with Dynamic Graphs
Paper#Graph Neural Networks, Log Analysis, Debugging🔬 Research|Analyzed: Jan 3, 2026 19:27•
Published: Dec 28, 2025 12:23
•1 min read
•ArXivAnalysis
This paper addresses the limitations of using large language models (LLMs) for debugging tabular logs, proposing a more flexible and scalable approach using dynamic graphs. The core idea is to represent the log data as a dynamic graph, allowing for efficient debugging with a simple Graph Neural Network (GNN). The paper's significance lies in its potential to reduce reliance on computationally expensive LLMs while maintaining or improving debugging performance.
Key Takeaways
- •Proposes GraphLogDebugger, a framework for debugging tabular logs using dynamic graphs.
- •Constructs heterogeneous nodes for objects and events and connects them with edges to represent the system as an evolving dynamic graph.
- •Demonstrates that a simple dynamic GNN can outperform LLMs in debugging tabular logs.
- •Offers a more flexible and scalable alternative to LLM-based approaches.
Reference / Citation
View Original"A simple dynamic Graph Neural Network (GNN) is representative enough to outperform LLMs in debugging tabular log."