Attention-Based Semantic Log Parsing and Core Dump Analysis for Zero-Shot Error Classification
Keywords:
Log parsing, Core dump analysis, Zero-shot classification, Attention mechanism, Error classificationAbstract
Modern software systems generate extensive logs and core dumps that contain critical diagnostic evidence for error classification. However, traditional parsing pipelines depend on handcrafted rules and labeled failure examples, limiting their applicability to unseen error categories. This research presents an attention-based semantic log parsing and core dump analysis framework for zero-shot error classification. The method first transforms unstructured log messages into semantic embeddings using a neural parser that learns template boundaries and event relationships without manual regular expressions. In parallel, core dump files are analyzed to extract memory state features, stack trace signatures, and register-level anomaly indicators. These two modalities are fused through a multi-head attention mechanism that aligns textual log semantics with binary fault artifacts. A zero-shot classifier then assigns error categories by matching fused representations to semantic prototypes, enabling classification of previously unseen faults without retraining. Experiments were conducted on a benchmark corpus of application logs and corresponding core dumps spanning memory errors, null pointer dereference, segmentation fault, and concurrency deadlock categories. The full framework achieved strong precision, recall, and balanced classification quality across all error types, with the highest scores for segmentation fault and memory errors. Ablation analysis confirmed that both semantic attention and core dump features contribute substantially; removing either modality degraded classification quality, and removing both produced the lowest performance. The framework also maintained competitive inference latency, making it suitable for near-real-time diagnostic support. These findings indicate that attention-based multimodal fusion offers a practical path toward robust zero-shot error classification in complex software environments. Future work will extend the approach to streaming logs and heterogeneous core dump formats.References
1. E. Karlsen, X. Luo, N. Zincir-Heywood, and M. Heywood, “Benchmarking large language models for log analysis, security, and interpretation,” J. Netw. Syst. Manag., vol. 32, no. 3, p. 59, 2024.
2. Z. Ma, D. J. Kim, and T. H. Chen, “LibreLog: Accurate and efficient unsupervised log parsing using open-source large language models,” arXiv preprint arXiv:2408.01585, 2024.
3. W. Hong, Y. Wu, L. Zhang, C. Duan, P. Xiao, M. He et al., “Cslparser: A collaborative framework using small and large language models for log parsing,” in *2025 IEEE 36th International Symposium on Software Reliability Engineering (ISSRE)*, 2025, pp. 61–72.
4. S. He, P. He, Z. Chen, T. Yang, Y. Su, and M. R. Lyu, “A survey on automated log analysis for reliability engineering,” ACM Comput. Surv., vol. 54, no. 6, pp. 1–37, 2021.
5. Z. Jiang, J. Liu, J. Huang, Y. Li, Y. Huo, J. Gu et al., “A large-scale evaluation for log parsing techniques: How far are we?,” in *Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis*, 2024, pp. 223–234.
6. C. Zhang, W. Xu, J. Liu, L. Zhang, G. Liu, J. Guan et al., “Logbase: A large-scale benchmark for semantic log parsing,” Proc. ACM Softw. Eng., vol. 2, no. ISSTA, pp. 2091–2112, 2025.
7. M. Astekin, M. Hort, and L. Moonen, “A comparative study on large language models for log parsing,” in *Proceedings of the 18th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement*, 2024, pp. 234–244.
8. S. Saibharath and G. Geethakumari, “Cloud forensics: Evidence collection and preliminary analysis,” in 2015 IEEE International Advance Computing Conference (IACC), 2015, pp. 464–467.
9. A. Aljaedi, D. Lindskog, P. Zavarsky, R. Ruhl, and F. Almari, “Comparative analysis of volatile memory forensics: live response vs. memory imaging,” in *2011 IEEE Third International Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third International Conference on Social Computing*, 2011, pp. 1253–1258.
10. S. M. Hejazi, M. Debbabi, and C. Talhi, “Automated windows memory file extraction for cyber forensics investigation,” J. Digit. Forensic Pract., vol. 2, no. 3, pp. 117–131, 2008.
11. V. H. Le and H. Zhang, “Log parsing with prompt-based few-shot learning,” in 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE), 2023, pp. 2438–2449.
12. S. Yu, Y. Wu, Z. Li, P. He, N. Chen, and C. Liu, “Log parsing with generalization ability under new log types,” in *Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering*, 2023, pp. 425–437.
13. P. Saluja and M. Bhavsar, “KDCFI AI-based knowledge discovery framework for cloud forensic investigation,” J. Supercomput., vol. 81, no. 17, p. 1572, 2025.
14. P. He, J. Zhu, Z. Zheng, and M. R. Lyu, “Drain: An online log parsing approach with fixed depth tree,” in 2017 IEEE International Conference on Web Services (ICWS), 2017, pp. 33–40.
15. D. Tang and L. Lindholm, “Hierarchical Self-Supervised Learning for Semantic Embeddings of Crash Dumps,” LU-CS-EX, 2026.
16. J. Zhang, H. Gao, Y. Liang et al., “Scale-Separated Conditioning for Style-Encoder-Free Diffusion Stylization,” arXiv preprint arXiv:2608.19719, 2026.
17. A. Schuster, “Searching for processes and threads in Microsoft Windows memory dumps,” Digit. Investig., vol. 3, pp. 10–16, 2006.
18. Z. Li, Q. Fu, Z. Huang, J. Yu, Y. Li, Y. Lai, and Y. Ma, “Revisiting log parsing: The present, the future, and the uncertainties,” IEEE Trans. Rel., vol. 73, no. 3, pp. 1459–1472, 2024.
19. T. Liao, X. Fang, Y. Feng, and S. Wang, “Reliable and Efficient Decision-Making for Large Language Model Agents through Uncertainty-Guided Adaptive Reasoning,” 2026. [Online]. Available: SSRN 6844659.
20. S. Gholamian and P. A. Ward, “A comprehensive survey of logging in software: From logging statements automation to log mining and analysis,” arXiv preprint arXiv:2110.12489, 2021.
21. J. Xu, R. Yang, Y. Huo, C. Zhang, and P. He, “Divlog: Log parsing with prompt enhanced in-context learning,” in *Proceedings of the IEEE/ACM 46th International Conference on Software Engineering*, 2024, pp. 1–12.
22. J. Bellizzi, “Responding to stealthy attacks on android using timely-captured memory dumps,” 2023.
23. Y. Li, “Ai-augmented automated triage for system error reporting using log and memory dump analysis,” in *2026 7th International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE)*, 2026, pp. 572–577.
24. D. Weeratunge, X. Zhang, and S. Jagannathan, “Analyzing multicore dumps to facilitate concurrency bug reproduction,” in *Proceedings of the fifteenth International Conference on Architectural Support for Programming Languages and Operating Systems*, 2010, pp. 155–166.
25. A. Das, P. Jiang, D. Zhu, B. Poczos, and J. Lugo-Martinez, “State-space architectures for scalable diffusion-based 3D molecule generation,” in *NeurIPS 2025 Workshop on Structured Probabilistic Inference & Generative Modeling*, 2025.
26. V. H. Le and H. Zhang, “Log parsing: How far can chatgpt go?,” in *2023 38th IEEE/ACM International Conference on Automated Software Engineering (ASE)*, 2023, pp. 1699–1704.
27. Y. Xiao, V. H. Le, and H. Zhang, “free: Towards more practical log parsing with large language models,” in *Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering*, 2024, pp. 153–165.
28. W. Cui, M. Peinado, S. K. Cha, Y. Fratantonio, and V. P. Kemerlis, “Retracer: Triaging crashes by reverse execution from partial memory dumps,” in Proceedings of the 38th International Conference on Software Engineering, 2016, pp. 820–831.
29. B. Dolan-Gavitt, “Forensic analysis of the Windows registry in memory,” Digit. Investig., vol. 5, pp. S26–S32, 2008.
30. D. Yuan, H. Mai, W. Xiong, L. Tan, Y. Zhou, and S. Pasupathy, “Sherlog: error diagnosis by connecting clues from run-time logs,” in *Proceedings of the fifteenth International Conference on Architectural Support for Programming Languages and Operating Systems*, 2010, pp. 143–154.
31. R. Liu, W. Guo, T. Liao, and X. Fang, “Uncertainty-Aware Adaptive Debate for Robust and Efficient Large Language Model Reasoning,” 2026.
32. J. Ma, Y. Liu, H. Wan, and G. Sun, “Automatic parsing and utilization of system log features in log analysis: A survey,” Appl. Sci., vol. 13, no. 8, p. 4930, 2023.
33. Y. Xiao, V. H. Le, and H. Zhang, “Stronger, cheaper and demonstration-free log parsing with LLMs,” arXiv preprint arXiv:2406.06156, 2024.
34. C. Zhang, W. Xu, J. Liu, L. Zhang, G. Liu, J. Guan et al., “SemanticLog: Towards Effective and Efficient Large-Scale Semantic Log Parsing,” IEEE Trans. Softw. Eng., 2025.
35. J. H. Lang and T. Schreck, “Leveraging LLMs for memory forensics: a comparative analysis of malware detection,” Digit. Threats: Res. Pract., vol. 6, no. 4, pp. 1–20, 2025.
