Share story
Science AI

Preprint tests execution checks before AI repairs software vulnerabilities

Researchers report in a preprint that an AI framework resolved 69.74% of software vulnerabilities end-to-end, with a 12.27% total failure rate. The system requires execution-based confirmation of exploitability before attempting a repair. It combines vulnerability detection, validation and iterative repair, using a shared representation of Java, Python and C++ code to support analysis across languages. Detection accuracy within individual languages reached 89.84-92.02%; zero-shot cross-language F1, a measure combining precision and recall, reached 74.43-80.12%. Disabling validation increased unnecessary repairs by 131.7%, demonstrating the operational value of checking exploitability before changing code.
Sources
cs.LG updates on arXiv.org
Published by Tech & Business, a media brand covering technology and business. This story was sourced from cs.LG updates on arXiv.org and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
Capital
Capital

IRACE Digital Bank acquires blockchain startup Trrue

IRACE Digital Bank, formerly Fundbank, has acquired Irish blockchain startup Trrue in a transaction valued at $11.8 million, CoinTrust reports. Dealroom reported a different value of $11.2 million. The acquisition brings Trrue's ...

CoinTrust
AI
AI

AWS releases open-source model for choosing AI workflow actions

Amazon Web Services has released Strands Decider 2B, an open-source model that selects among predefined options instead of generating open-ended text. It can choose an AI agent's next tool, route customer requests, score outputs a...

qatarliving.com
AI Science
AI Science

AI agents develop physical model from quantum-material observations

Researchers report in an arXiv preprint that their AI Theorist system autonomously developed a physical model explaining previously unpublished experimental observations in α-RuCl₃, a candidate material for realizing a Kitaev quan...

cs.LG updates on arXiv.org
Security AI
Security AI

MIRROR preprint reports blocking message tampering between AI agents

Researchers report in an arXiv preprint that MIRROR, a defense against message tampering between AI agents, reduced attack success rates to 0% in tests below its route-compromise threshold, at 1x LLM token cost. Tests covered thre...

cs.CR updates on arXiv.org
AI Science
AI Science

Researchers automate viral plaque counting in dengue vaccine assays

Researchers demonstrated CellVision, a deep learning system that fully automates viral plaque counting in high-throughput vaccine assays, BIOENGINEER reported. The image analysis platform separates fused plaques and distinguishes ...

BIOENGINEER