arXiv cs.AI· Leo Betthauser, Yingqi Liu, Uday Shah, Jishang Wei, Sainyam Galhotra, Liang Gou, Vedant Dharnidharka, Hao Yang·· 3 小时前AI 评分36
Lev-3B：面向轻量级概率日志决策模型的数据配方
Lev-3B: A Data Recipe for Compact Probabilistic Log Decision Models
AI 导读
研究者发布开源 3B 参数日志决策模型 Lev-3B，基于 Granite-4.1-3B-Base 构建，采用两阶段数据配方：先做不含日志数据的有界自然语言决策课程训练，再混入 35 个辅助任务的一般日志数据训练。
正文 · 原文
Abstract:Many log-analysis tasks involve choosing among known options, such as detecting anomalous sessions or identifying attack techniques, and benefit from calibrated probabilities. We present Lev-3B, an open 3B-parameter decision model built on Granite-4.1-3B-Base with a shared pointer readout, and study its training recipe. The recipe has two key parts. First, training follows a curriculum: the model learns bounded natural-language decisions, using no log data, before it sees any logs. Second, the log stage mixes the target tasks with general log data spanning 35 auxiliary tasks unrelated to either evaluation task. We evaluate one shared checkpoint on sources seen in training and on sources excluded from training, calibration and model selection. The two-stage recipe transfers substantially better than training on log data alone. Adding general log data improves performance on excluded sources. At a fraction of their serving cost, Lev-3B outperforms prompted frontier models on anomaly detection, including on an excluded logging system, but trails them on attack-technique classification. Its probabilities are well calibrated on familiar sources, but on the excluded logging system it confidently misses anomalies. We release the checkpoint, recipe, data manifests, and prediction-level artifacts.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.38915 [cs.LG] |
| (or arXiv:2610.38915v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38915 arXiv-issued DOI via DataCite |
Submission history
From: Sainyam Galhotra [view email]
[v1]
Wed, 28 Oct 2026 21:06:44 UTC (58 KB)
来源：arXiv cs.AI · arxiv.org