Primus: A Pioneering Collection of Open-Source Datasets for Cybersecurity LLM Training

ACL ARR 2025 May Submission4626 Authors

Abstract: Large Language Models (LLMs) have shown remarkable advancements in specialized fields such as finance, law, and medicine. However, in cybersecurity, we have noticed a lack of open-source datasets, with a particular lack of high-quality cybersecurity pretraining corpora, even though much research indicates that LLMs acquire their knowledge during pretraining. To address this, we present a comprehensive suite of datasets covering all major training stages, including pretraining, instruction fine-tuning, and reasoning distillation with cybersecurity-specific self-reflection data. Extensive ablation studies demonstrate their effectiveness on public cybersecurity benchmarks. In particular, continued pre-training on our dataset yields a **15.9%** improvement in the aggregate score, while reasoning distillation leads to a **15.8%** gain in security certification (CISSP). We will release all datasets and trained cybersecurity LLMs under the ODC-BY and MIT licenses to encourage further research in the community.

Paper Type: Long

Research Area: Resources and Evaluation

Research Area Keywords: pre-training, continual learning, fine-tuning, reasoning, few-shot QA, chain-of-thought, applications

Contribution Types: Model analysis & interpretability, Publicly available software and/or pre-trained models, Data resources, Data analysis

Languages Studied: English

Submission Number: 4626