LoRA-tuned tiny LLMs achieve 97.76-98.83% accuracy for Thunderbird log anomaly detection, beating a LogBERT full fine-tune baseline by 18-19 percentage points.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LogTinyLLM: Tiny Large Language Models Based Contextual Log Anomaly Detection
LoRA-tuned tiny LLMs achieve 97.76-98.83% accuracy for Thunderbird log anomaly detection, beating a LogBERT full fine-tune baseline by 18-19 percentage points.