Log Anomaly Detection on HDFS
99.4F1 ScoreCLAD
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CLADParser=false2026.04 | 99.4 | 99.06 | 99.74 | |
| NeuralLogParser=false2026.04 | 99.33 | 99.01 | 99.67 | |
| DeBERTa-v3Category=Fine-Tuned2026.04 | 98.9 | — | — | |
| RoBERTa-baseCategory=Fine-Tuned2026.04 | 98.2 | — | — | |
| BERT-baseCategory=Fine-Tuned2026.04 | 97.8 | — | — | |
| Drain + RFCategory=Traditional2026.04 | 95.1 | — | — | |
| Drain + RF (best)Category=Traditional2026.04 | 95.1 | — | — | |
| LogRobustParser=true2026.04 | 93.89 | 92.42 | 95.41 | |
| GPT-4 + SLCP (5-shot)Category=LLM (Prompt), Shot-count=5, Prompting strategy (SLCP)=true2026.04 | 93.8 | — | — | |
| Drain + LRCategory=Traditional2026.04 | 93.2 | — | — | |
| Spell + SVMCategory=Traditional2026.04 | 92.8 | — | — | |
| GPT-4 (5-shot)Category=LLM (Prompt), Shot-count=5, Prompting strategy (SLCP)=false2026.04 | 91.7 | — | — | |
| GPT-4 + SLCP (zero)Category=LLM (Prompt), Shot-count=0, Prompting strategy (SLCP)=true2026.04 | 91.2 | — | — | |
| CNNParser=true2026.04 | 90.16 | 89.32 | 91.01 | |
| GPT-4 (zero-shot)Category=LLM (Prompt), Shot-count=0, Prompting strategy (SLCP)=false2026.04 | 88.3 | — | — | |
| LLaMA-3 (zero-shot)Category=LLM (Prompt), Shot-count=0, Prompting strategy (SLCP)=false2026.04 | 84.1 | — | — | |
| GPT-3.5 (zero-shot)Category=LLM (Prompt), Shot-count=0, Prompting strategy (SLCP)=false2026.04 | 82.4 | — | — |