Transformer and Mamba models lead in automated log parsing research

A new research paper on arXiv evaluates four sequence modeling architectures for automated log parsing, finding that Transformers achieve the best performance with a mean relative edit distance of 0.111. Mamba models offer a competitive alternative with significantly lower computational costs. The study also indicates that character-level tokenization generally improves accuracy, while sequence length has minimal impact on Transformer performance. AI

IMPACT Provides practical guidance on selecting sequence modeling architectures for automated log parsing, highlighting Transformer and Mamba's effectiveness.

RANK_REASON Academic paper published on arXiv detailing empirical study of sequence-to-sequence models for log parsing. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Andriy Miranskyy
• arXiv
• LSTM
• Mamba
• Transformer

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →