dorsal/arxiv
View SchemaMicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning
| Authors | Jianbo Yu, Yixuan Li, Hai Xu, Kang Xu, Junjielong Xu, Zhijing Li, Pinjia He, Wanyuan Wang |
|---|---|
| Categories | |
| ArXiv ID | 2601.07005vv1 |
| URL | https://arxiv.org/abs/2601.07005 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Log parsing converts semi-structured logs into structured templates, forming a critical foundation for downstream analysis. Traditional syntax and semantic-based parsers often struggle with semantic variations in evolving logs and data scarcity stemming from their limited domain coverage. Recent large language model (LLM)-based parsers leverage in-context learning (ICL) to extract semantics from examples, demonstrating superior accuracy. However, LLM-based parsers face two main challenges: 1) underutilization of ICL capabilities, particularly in dynamic example selection and cross-domain generalization, leading to inconsistent performance; 2) time-consuming and costly LLM querying. To address these challenges, we present MicLog, the first progressive meta in-context learning (ProgMeta-ICL) log parsing framework that combines meta-learning with ICL on small open-source LLMs (i.e., Qwen-2.5-3B). Specifically, MicLog: i) enhances LLMs' ICL capability through a zero-shot to k-shot ProgMeta-ICL paradigm, employing weighted DBSCAN candidate sampling and enhanced BM25 demonstration selection; ii) accelerates parsing via a multi-level pre-query cache that dynamically matches and refines recently parsed templates. Evaluated on Loghub-2.0, MicLog achieves 10.3% higher parsing accuracy than the state-of-the-art parser while reducing parsing time by 42.4%.
{
"annotation_id": "9058b25a-9ff5-40d8-ad79-10d51bbb7385",
"date_created": "2026-02-17T05:53:08.667000Z",
"date_modified": "2026-02-17T05:53:08.667000Z",
"file_hash": "fed25b741fdf56be25dd516b2e584540594db16fd6848b8aa33af1fda81540c1",
"private": false,
"record": {
"abstract": "Log parsing converts semi-structured logs into structured templates, forming a critical foundation for downstream analysis. Traditional syntax and semantic-based parsers often struggle with semantic variations in evolving logs and data scarcity stemming from their limited domain coverage. Recent large language model (LLM)-based parsers leverage in-context learning (ICL) to extract semantics from examples, demonstrating superior accuracy. However, LLM-based parsers face two main challenges: 1) underutilization of ICL capabilities, particularly in dynamic example selection and cross-domain generalization, leading to inconsistent performance; 2) time-consuming and costly LLM querying. To address these challenges, we present MicLog, the first progressive meta in-context learning (ProgMeta-ICL) log parsing framework that combines meta-learning with ICL on small open-source LLMs (i.e., Qwen-2.5-3B). Specifically, MicLog: i) enhances LLMs\u0027 ICL capability through a zero-shot to k-shot ProgMeta-ICL paradigm, employing weighted DBSCAN candidate sampling and enhanced BM25 demonstration selection; ii) accelerates parsing via a multi-level pre-query cache that dynamically matches and refines recently parsed templates. Evaluated on Loghub-2.0, MicLog achieves 10.3% higher parsing accuracy than the state-of-the-art parser while reducing parsing time by 42.4%.",
"arxiv_id": "2601.07005",
"authors": [
"Jianbo Yu",
"Yixuan Li",
"Hai Xu",
"Kang Xu",
"Junjielong Xu",
"Zhijing Li",
"Pinjia He",
"Wanyuan Wang"
],
"categories": [
"cs.SE",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning",
"url": "https://arxiv.org/abs/2601.07005",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "a7e4e4b7-3ecd-49c8-a67a-d52e6537eeb1",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}