arXiv preprint ·
Found, but Not Fixed : From Software Vulnerability Discovery to Fixes with Iterative LLM Pipelines
- Published
- Source
- arXiv preprint, posted before peer review
- Original
- Read it at arXiv
- Topics
Abstract
Software vulnerabilities persist as a critical threat despite the deployment of sophisticated detection tools and AI assistants. While detection has advanced, the subsequent code repair phase remains a primary bottleneck; manual intervention is inherently slow, costly, and prone to introducing secondary regressions and issues. In this paper, we investigate whether an iterative, LLM-based pipeline can provide a robust solution for the automated repair of vulnerable functions in real-world code. We propose a generalized framework that leverages the reasoning capabilities of Large Language Models (LLMs) through a multi-stage feedback loop. We benchmark this approach against state-of-the-art models, including GPT-4o, GPT-5.1, and Gemini 3 Flash, demonstrating that such a pipeline can achieve high-quality code repair. To support the broader security community, we provide our evaluation framework, fine-tuned weights, and comprehensive experimental dataset on our website: https://sites.google.com/view/llm4cve
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Software Engineering
Attacker Diversity Reduction in Adaptive LLM Red Teaming
Designing Deceptive Model Organisms that Successfully Evade Auditing
ReconPotBench: Evaluating LLM Honeypots in Multi-Round Linux Reconnaissance
When Backdoors Can Be Switched Off: Suppression-Aware Backdoor Attacks on Vision-Language-Action Models
MASTA: A Feedback-Scheduled Multi-Agent System for End-to-End Tamarin Modeling and Analysis
Get the few papers worth your time in Software Engineering every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.