Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.17948 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917422171160576 |
|---|---|
| author | Jamwal, Parteek Shao, Minghao Chen, Boyuan Muthuvelan, Achyuta Subanya, Asini Ballo, Boubacar Satija, Kashish Shafey, Mariam Mahmoud, Mohamed Bouffi, Moncif Dahaji Wickramasinghe, Pasindu Goel, Siyona Sabbani, Yaakulya Hacid, Hakim Ndhlovu, Mthandazo Kafeza, Eleanna Rawat, Sanjay Shafique, Muhammad |
| author_facet | Jamwal, Parteek Shao, Minghao Chen, Boyuan Muthuvelan, Achyuta Subanya, Asini Ballo, Boubacar Satija, Kashish Shafey, Mariam Mahmoud, Mohamed Bouffi, Moncif Dahaji Wickramasinghe, Pasindu Goel, Siyona Sabbani, Yaakulya Hacid, Hakim Ndhlovu, Mthandazo Kafeza, Eleanna Rawat, Sanjay Shafique, Muhammad |
| contents | Large Language Models (LLMs) have demonstrated remarkable capabilities across various cybersecurity tasks, including vulnerability classification, detection, and patching. However, their potential in automated vulnerability report documentation and analysis remains underexplored. We present RAVEN (Retrieval Augmented Vulnerability Exploration Network), a framework leveraging LLM agents and Retrieval Augmented Generation (RAG) to synthesize comprehensive vulnerability analysis reports. Given vulnerable source code, RAVEN generates reports following the Google Project Zero Root Cause Analysis template. The framework uses four modules: an Explorer agent for vulnerability identification, a RAG engine retrieving relevant knowledge from curated databases including Google Project Zero reports and CWE entries, an Analyst agent for impact and exploitation assessment, and a Reporter agent for structured report generation. To ensure quality, RAVEN includes a task specific LLM Judge evaluating reports across structural integrity, ground truth alignment, code reasoning quality, and remediation quality. We evaluate RAVEN on 105 vulnerable code samples covering 15 CWE types from the NIST-SARD dataset. Results show an average quality score of 54.21%, supporting the effectiveness of our approach for automated vulnerability documentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_17948 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RAVEN: Retrieval-Augmented Vulnerability Exploration Network for Memory Corruption Analysis in User Code and Binary Programs Jamwal, Parteek Shao, Minghao Chen, Boyuan Muthuvelan, Achyuta Subanya, Asini Ballo, Boubacar Satija, Kashish Shafey, Mariam Mahmoud, Mohamed Bouffi, Moncif Dahaji Wickramasinghe, Pasindu Goel, Siyona Sabbani, Yaakulya Hacid, Hakim Ndhlovu, Mthandazo Kafeza, Eleanna Rawat, Sanjay Shafique, Muhammad Cryptography and Security Artificial Intelligence Multiagent Systems Large Language Models (LLMs) have demonstrated remarkable capabilities across various cybersecurity tasks, including vulnerability classification, detection, and patching. However, their potential in automated vulnerability report documentation and analysis remains underexplored. We present RAVEN (Retrieval Augmented Vulnerability Exploration Network), a framework leveraging LLM agents and Retrieval Augmented Generation (RAG) to synthesize comprehensive vulnerability analysis reports. Given vulnerable source code, RAVEN generates reports following the Google Project Zero Root Cause Analysis template. The framework uses four modules: an Explorer agent for vulnerability identification, a RAG engine retrieving relevant knowledge from curated databases including Google Project Zero reports and CWE entries, an Analyst agent for impact and exploitation assessment, and a Reporter agent for structured report generation. To ensure quality, RAVEN includes a task specific LLM Judge evaluating reports across structural integrity, ground truth alignment, code reasoning quality, and remediation quality. We evaluate RAVEN on 105 vulnerable code samples covering 15 CWE types from the NIST-SARD dataset. Results show an average quality score of 54.21%, supporting the effectiveness of our approach for automated vulnerability documentation. |
| title | RAVEN: Retrieval-Augmented Vulnerability Exploration Network for Memory Corruption Analysis in User Code and Binary Programs |
| topic | Cryptography and Security Artificial Intelligence Multiagent Systems |
| url | https://arxiv.org/abs/2604.17948 |