Researchers have developed a method to automate the mapping of cloud security controls to technical metrics using domain-adapted Sentence Transformer models. By creating a training corpus from European security standards and technical metrics, and expanding it through back-translation and LLM-based paraphrasing, they achieved significant performance gains. The fine-tuned models outperformed zero-shot baselines, with one model showing up to a 23 nDCG@10 point improvement on the control-to-metric task and another reaching 0.870 nDCG@10 on the cross-standard control task. The study highlights the critical role of in-domain training data for enhancing model performance in these specific applications. AI
IMPACT This research demonstrates a practical application of LLMs for automating complex compliance tasks in cloud security, potentially reducing manual effort and improving accuracy.
RANK_REASON The cluster contains an academic paper detailing a new methodology and experimental results.
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Europe
- Gotit.pub
- Hugging Face
- multi-qa-mpnet-dot-v1
- ScienceCast
- Sentence Transformer models
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
