Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908795806941184 |
|---|---|
| author | Yang, Zhuoran Li, Ed He, Jianliang Priyanshu, Aman Saglam, Baturay Kassianik, Paul Weerawardhena, Sajana Vellore, Anu Nelson, Blaine Javidnia, Neusha Goldblatt, Arthur Burch, Fraser Zohary, Avi Eisenman, Assaf Sabbaghi, Mahdi Vijay, Supriti Dharssi, Rahim Kedia, Dhruv Oshiba, Kojin Singer, Yaron Karbasi, Amin |
| author_facet | Yang, Zhuoran Li, Ed He, Jianliang Priyanshu, Aman Saglam, Baturay Kassianik, Paul Weerawardhena, Sajana Vellore, Anu Nelson, Blaine Javidnia, Neusha Goldblatt, Arthur Burch, Fraser Zohary, Avi Eisenman, Assaf Sabbaghi, Mahdi Vijay, Supriti Dharssi, Rahim Kedia, Dhruv Oshiba, Kojin Singer, Yaron Karbasi, Amin |
| contents | We present Foundation-Sec-8B-Reasoning, the first open-source native reasoning model for cybersecurity. Built upon our previously released Foundation-Sec-8B base model (derived from Llama-3.1-8B-Base), the model is trained through a two-stage process combining supervised fine-tuning (SFT) and reinforcement learning from verifiable rewards (RLVR). Our training leverages proprietary reasoning data spanning cybersecurity analysis, instruction-following, and mathematical reasoning. Evaluation across 10 cybersecurity benchmarks and 10 general-purpose benchmarks demonstrates performance competitive with significantly larger models on cybersecurity tasks while maintaining strong general capabilities. The model shows effective generalization on multi-hop reasoning tasks and strong safety performance when deployed with appropriate system prompts and guardrails. This work demonstrates that domain-specialized reasoning models can achieve strong performance on specialized tasks while maintaining broad general capabilities. We release the model publicly at https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21051 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report Yang, Zhuoran Li, Ed He, Jianliang Priyanshu, Aman Saglam, Baturay Kassianik, Paul Weerawardhena, Sajana Vellore, Anu Nelson, Blaine Javidnia, Neusha Goldblatt, Arthur Burch, Fraser Zohary, Avi Eisenman, Assaf Sabbaghi, Mahdi Vijay, Supriti Dharssi, Rahim Kedia, Dhruv Oshiba, Kojin Singer, Yaron Karbasi, Amin Artificial Intelligence Cryptography and Security Machine Learning We present Foundation-Sec-8B-Reasoning, the first open-source native reasoning model for cybersecurity. Built upon our previously released Foundation-Sec-8B base model (derived from Llama-3.1-8B-Base), the model is trained through a two-stage process combining supervised fine-tuning (SFT) and reinforcement learning from verifiable rewards (RLVR). Our training leverages proprietary reasoning data spanning cybersecurity analysis, instruction-following, and mathematical reasoning. Evaluation across 10 cybersecurity benchmarks and 10 general-purpose benchmarks demonstrates performance competitive with significantly larger models on cybersecurity tasks while maintaining strong general capabilities. The model shows effective generalization on multi-hop reasoning tasks and strong safety performance when deployed with appropriate system prompts and guardrails. This work demonstrates that domain-specialized reasoning models can achieve strong performance on specialized tasks while maintaining broad general capabilities. We release the model publicly at https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Reasoning. |
| title | Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report |
| topic | Artificial Intelligence Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2601.21051 |