Research map / Large language models / Language, safety and interpretability
MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning
Jianbo Yu, Yixuan Li, Hai Xu, Kang Xu, Junjielong Xu, Zhijing Li, Pinjia He, Wanyuan Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add progressive meta in-context training (ablation) | MicLog · Loghub-2.0 | PTA: 60.2 -> 95.3 | Improved |
| add progressive meta in-context training (ablation) | MicLog · Loghub-2.0 | PA: 87.3 -> 97.6 | Improved |
| use weighted DBSCAN instead of random sampling (ablation) | MicLog · Loghub-2.0 | PTA: 76.8 -> 95.3 | Improved |
| use weighted DBSCAN instead of k-means (ablation) | MicLog · Loghub-2.0 | PTA: 92.0 -> 95.3 | Improved |
| use BM25 selection instead of random (ablation) | MicLog · Loghub-2.0 | PTA: 79.1 -> 95.3 | Improved |
| use multi-level cache instead of LILAC cache (ablation) | MicLog · Loghub-2.0 | PA: 97.6 -> 97.6 | No clear effect |
| multi-level cache | MicLog · Loghub-2.0 | cache time (s): 376.5 -> 63.5 (LILAC vs MicLog) | Improved |
| train 5 epochs instead of 1 (ablation) | MicLog · Loghub-2.0 | PA: 96.3 -> 97.6 | Improved |
| progressive 0-to-5 shot meta-training vs 5-shot only (ablation) | MicLog · Loghub-2.0 | PTA: 87.5 -> 95.3 | Improved |
| meta-train on Proxifier only (ablation) | MicLog · Loghub-2.0 | PA: 97.6 -> 95.8 | No clear effect |
Closest papers by meaning
- Filling Memory Gaps: Enhancing Continual Semantic Parsing via SQL Syntax Variance-Guided LLMs Without Real Data Replay (AAAI 2025 · similarity 0.85)
- metabench - A Sparse Benchmark of Reasoning and Knowledge in Large Language Models (ICLR 2025 · similarity 0.85)
- The Missing Alignment Link of In-context Learning on Sequences (ICML 2025 · similarity 0.85)
- In-Context Language Learning: Architectures and Algorithms (ICML 2024 · similarity 0.85)
- Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAG (ICLR 2025 · similarity 0.84)
- MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning? (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
