[2603.19329] Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification

[2603.19329] Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification

arXiv - AI 4 min read

About this article

Abstract page for arXiv paper 2603.19329: Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification

Computer Science > Software Engineering arXiv:2603.19329 (cs) [Submitted on 18 Mar 2026] Title:Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification Authors:Zenan Li, Ziran Yang, Deyuan (Mike)He, Haoyu Zhao, Andrew Zhao, Shange Tang, Kaiyu Yang, Aarti Gupta, Zhendong Su, Chi Jin View a PDF of the paper titled Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification, by Zenan Li and 9 other authors View PDF HTML (experimental) Abstract:Large language models (LLMs) can generate plausible code but offer limited guarantees of correctness. Formally verifying that implementations satisfy specifications requires constructing machine-checkable proofs, a task that remains beyond current automation. We propose a hierarchical proof search framework for automated code verification in Lean~4 that decomposes complex verification goals into structurally simpler subgoals before attempting tactic-level proving. Central to our approach is a principled decomposition score that combines constructive justification with structural effectiveness. Crucially, this score serves as both the training reward and the inference-time ranking criterion, ensuring strict alignment between optimization and deployment. We train Goedel-Code-Prover-8B, a single unified policy for both decomposition and completion, via supervised initialization followed by hybrid reinforcement learning, where a continuous decomposition reward drives planning ex...

Originally published on March 23, 2026. Curated by AI News.

Related Articles

Llms

OpenClaw security checklist: practical safeguards for AI agents

Here is one of the better quality guides on the ensuring safety when deploying OpenClaw: https://chatgptguide.ai/openclaw-security-checkl...

Reddit - Artificial Intelligence · 1 min ·
I let Gemini in Google Maps plan my day and it went surprisingly well | The Verge
Llms

I let Gemini in Google Maps plan my day and it went surprisingly well | The Verge

Gemini in Google Maps is a surprisingly useful way to explore new territory.

The Verge - AI · 11 min ·
Llms

The person who replaces you probably won't be AI. It'll be someone from the next department over who learned to use it - opinion/discussion

I'm a strategy person by background. Two years ago I'd write a recommendation and hand it to a product team. Now.. I describe what I want...

Reddit - Artificial Intelligence · 1 min ·
Block Resets Management With AI As Cash App Adds Installment Transfers
Llms

Block Resets Management With AI As Cash App Adds Installment Transfers

Block (NYSE:XYZ) plans a permanent organizational overhaul that replaces many middle management roles with AI-driven models to create fla...

AI Tools & Products · 5 min ·
More in Llms: This Week Guide Trending

No comments

No comments yet. Be the first to comment!

Stay updated with AI News

Get the latest news, tools, and insights delivered to your inbox.

Daily or weekly digest • Unsubscribe anytime