Dissect-and-Restore: AI-based Code Verification with Transient Refactoring

2025-10-29

AI Summary

Formal verification faces high barriers to adoption due to its reliance on expert knowledge and extensive manual annotation. To address this, we propose a “decompose-recombine” verification paradigm: complex program logic is modularly decomposed into independently verifiable subcomponents; AI-driven transient code refactoring automatically solves each subproblem and assembles the corresponding correctness proofs; and natural language interaction guides the verification process. Our approach integrates formal verification, modular code restructuring, AI-based reasoning, and natural language understanding to achieve automated, low-code verification. Experimental evaluation demonstrates an 86% verification success rate on standard benchmarks—18 percentage points higher than baseline methods. In complex scenarios, success rises significantly from 30% to 69%; when integrated with mainstream proof frameworks, performance reaches 87%.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Knowledge Representation and Reasoning: Automated Reasoning and Theorem Proving

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Formal verification is increasingly recognized as a critical foundation for building reliable software systems. However, the need for specialized expertise to write precise specifications, navigate complex proof obligations, and learn annotations often makes verification an order of magnitude more expensive than implementation. While modern AI systems can recognize patterns in mathematical proofs and interpret natural language, effectively integrating them into the formal verification process remains an open challenge. We present Prometheus, a novel AI-assisted system that facilitates automated code verification with current AI capabilities in conjunction with modular software engineering principles (e.g., modular refactoring). Our approach begins by decomposing complex program logic, such as nested loops, into smaller, verifiable components. Once verified, these components are recomposed to construct a proof of the original program. This decomposition-recomposition workflow is non-trivial. Prometheus addresses this by guiding the proof search through structured decomposition of complex lemmas into smaller, verifiable sub-lemmas. When automated tools are insufficient, users can provide lightweight natural language guidance to steer the proof process effectively. Our evaluation demonstrates that transiently applying modular restructuring to the code substantially improves the AI's effectiveness in verifying individual components. This approach successfully verifies 86% of tasks in our curated dataset, compared to 68% for the baseline. Gains are more pronounced with increasing specification complexity, improving from 30% to 69%, and when integrating proof outlines for complex programs, from 25% to 87%.

Problem

• Automating formal verification of complex software systems
• Reducing expertise barriers in writing specifications and proofs
• Integrating AI with modular decomposition for scalable verification

Innovation

• Decomposes complex programs into verifiable components
• Recomposes verified components to prove original program
• Uses natural language guidance to steer proof process