Rango is an AI-powered tool that automatically generates formal proofs for software verification, reducing the expertise needed to ensure software security.

• Uses adaptive retrieval-augmented generation to find relevant premises for proof synthesis
• Achieves state-of-the-art performance in automated Coq theorem proving
• Employs a novel approach that dynamically adjusts retrieval strategies based on proof context
• Significantly reduces the manual effort required for formal verification

This research advances software security by making formal verification more accessible, helping engineering teams build more reliable and secure systems without extensive proof assistant expertise.

Rango: Adaptive Retrieval-Augmented Proving for Automated Software Verification