English(EN) Retrieval Supersedes Scale: Efficient Log Level Prediction via Multiplex Socio-Technical Context

LLM利用社会技术背景改进日志级别预测

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-05 05:00

研究人员开发了一种新的基于检索的方法，用于大型语言模型（LLM）预测软件开发中的日志级别，旨在提高效率和准确性。该方法利用社会技术背景，同时考虑代码功能和开发者所有权，而不是依赖随机示例。所提出的技术显著优于随机基线和现有的监督预测器，在多个Java项目中实现了更高的AUC分数。 AI

影响通过提高日志分析的准确性和效率，增强了LLM在软件工程中的效用。

排序理由该集群包含一篇学术论文，详细介绍了使用LLM进行日志级别预测的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 26 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了使用LLM进行日志级别预测的新方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, product; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.LG TIER_1 English(EN) · Youssef Esseddiq Ouatiti, Mohammed Sayagh, Bram Adams, Ahmed E. Hassan · 2026-11-05 05:00

检索超越规模：通过多路社会技术背景实现高效日志级别预测

arXiv:2611.05901v1 Announce Type: cross Abstract: Developers insert logging statements in source code to record its execution, and the resulting logs are essential for detecting and troubleshooting system anomalies. A key part of these statements is the log level (e.g., DEBUG, IN…