martin-wey/R2Vul
R2Vul: Learning to Reason about Software Vulnerabilities with Reinforcement Learning and Structured Reasoning Distillation
This project offers a way to analyze software code and identify potential vulnerabilities. It takes code samples as input and outputs detailed reasoning about where security flaws might exist. Software security researchers and developers focusing on code quality would find this valuable for understanding and improving vulnerability detection.
No commits in the last 6 months.
Use this if you are a researcher or developer who wants to train and evaluate models for identifying and reasoning about software vulnerabilities.
Not ideal if you are looking for a ready-to-use tool to scan production code for vulnerabilities without any model training or setup.
Stars
15
Forks
1
Language
Python
License
MIT
Category
Last pushed
Aug 05, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/martin-wey/R2Vul"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lin-tan/clm
For our ICSE23 paper "Impact of Code Language Models on Automated Program Repair" by Nan Jiang,...
waroad/losver
Source Code for LOSVER: Line-Level Modifiability Signal-Guided Vulnerability Detection and Classification
thanhlecongg/Invalidator
Invalidator: Automated Patch Correctness Assessment via Semantic and Syntactic Reasoning (IEEE TSE)
nghiempt/llm-analysis-privacy-policy
Unveiling Discrepancies in Android App Data Safety Declarations and Privacy Policies: An...
garghub/VulnerabilityCouplingMutants
On the Coupling between Vulnerabilities and LLM-generated Mutants: A Study on Vul4J dataset, The...