← All papers  /  Nov 19, 2026
Agents

A Tensor-Algebraic No-Go Theorem for High-Dimensional Photonic Greenberger--Horne--Zeilinger States Discovered with AlphaProof Nexus

First page
A Tensor-Algebraic No-Go Theorem for High-Dimensional Photonic Greenberger--Horne--Zeilinger States Discovered with AlphaProof Nexus
The curator’s take

Mario Krenn at the University of Tübingen, with Moritz Firsching and George Tsoukalas at Google DeepMind and colleagues, prove that high-dimensional photonic GHZ states cannot be built with minimal linear-optics resources, using a proof found by AlphaProof Nexus, an agentic system that searches for formal proofs in Lean.

Ask this paper

Key points
01

The physics question. Photonic GHZ states from probabilistic pair sources map to edge-coloured, complex-weighted graphs. Generating an N-photon, D-dimensional GHZ state is equivalent to building a monochromatic quantum graph.

02

The theorem. No complex-weighted monochromatic quantum graph exists for any even N of at least 4 when D is at least N. This is the first unconditional proof of a general case of a question the authors raised in 2017.

03

How the proof works. It builds a weighted sum of amplitudes whose coefficients factor vertex by vertex. A row-column imbalance makes the sum zero for the graph but nonzero for the GHZ target, so the two cannot be equal.

04

AI and interpretation. AlphaProof Nexus combines frontier language models with Lean compiler feedback and found the formal proof. The authors then give a physical interpretation of the proof.

Abstract

Photonic GHZ states generated with probabilistic pair sources can be described by edge-coloured, complex-weighted graphs. In this representation, conditioning on an event with one photon in every detector turns the output amplitudes into coherent sums over perfect matchings, and generating an \(N\)-photon \(D\)-dimensional GHZ state is equivalent to constructing a monochromatic quantum graph. Here, we prove that no complex-weighted monochromatic quantum graph exists for any even \(N\ge4\) when \(D\ge N\). The proof, discovered through AI-assisted formal proof search in Lean using AlphaProof Nexus, constructs a weighted sum \(\mathcal F\) of amplitudes whose coefficients factor vertex by vertex. Summing over the colours at one fixed vertex reduces the problem to an \((N-1)\times N\) matrix of effective edge weights. For the graph amplitudes \(ω\) and GHZ target \(Δ\), the row--column imbalance yields \(\mathcal F(ω)=0\) but \(\mathcal F(Δ)\neq0\), contradicting the required equality \(ω=Δ\). Finally, we give a physical interpretation of the AI-discovered mathematical proof.

Every Monday
Get next week’s papers.
Subscribe on Substack