Saved in:
Bibliographic Details
Main Authors: An, Chenyang, Ye, Qihao, Pan, Minghao, Zhang, Jiayaun
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.24021
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910248003960832
author An, Chenyang
Ye, Qihao
Pan, Minghao
Zhang, Jiayaun
author_facet An, Chenyang
Ye, Qihao
Pan, Minghao
Zhang, Jiayaun
contents We present \textbf{QED}, an open-source multi-agent system that turns human-provided research questions into complete mathematical proofs without further human guidance. Its pipeline is designed to overcome common failures of single-query proof generation by separating planning, proving, and verification: a decomposition agent structures the proof search, prover agents generate candidate arguments, and verifier agents check correctness. In collaboration with domain experts, we evaluated QED on 18 research-level projects of varying difficulty. QED produced five original works across algebraic geometry, fluid PDEs, probability, and inverse problems. Expert assessments regard these works as solid specialized research contributions, with three comparable in difficulty and scope to work commonly published in established specialist mathematics venues. QED is released at https://github.com/proofQED/QED.
format Preprint
id arxiv_https___arxiv_org_abs_2604_24021
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle QED: An Open-Source Multi-Agent System for Generating Mathematical Proofs on Open Problems
An, Chenyang
Ye, Qihao
Pan, Minghao
Zhang, Jiayaun
Artificial Intelligence
Analysis of PDEs
We present \textbf{QED}, an open-source multi-agent system that turns human-provided research questions into complete mathematical proofs without further human guidance. Its pipeline is designed to overcome common failures of single-query proof generation by separating planning, proving, and verification: a decomposition agent structures the proof search, prover agents generate candidate arguments, and verifier agents check correctness. In collaboration with domain experts, we evaluated QED on 18 research-level projects of varying difficulty. QED produced five original works across algebraic geometry, fluid PDEs, probability, and inverse problems. Expert assessments regard these works as solid specialized research contributions, with three comparable in difficulty and scope to work commonly published in established specialist mathematics venues. QED is released at https://github.com/proofQED/QED.
title QED: An Open-Source Multi-Agent System for Generating Mathematical Proofs on Open Problems
topic Artificial Intelligence
Analysis of PDEs
url https://arxiv.org/abs/2604.24021