Robust Bell Nonlocality from Gottesman–Kitaev–Preskill States
Xiaotian Yang, Santiago Zamora, Rafael Chaves, Ulrik L. Andersen, Jonatan Bohr Brask, A. de Oliveira Junior · 2026
Read Robust Bell Nonlocality from Gottesman–Kitaev–Preskill States by Xiaotian Yang, Santiago Zamora, Rafael Chaves, Ulrik L. Andersen, Jonatan Bohr Brask, A. de Oliveira Junior (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16189.txt.
Introduction
Xiaotian Yang and coauthors test if a grid code can make Bell tests work with plain reads. Their 2026 preprint deals with Bell tests that use wave reads. Wave reads in such tests face tight caps on what they can show. The team asks if GKP code can lift those caps. GKP code stores a bit in a wave like state on a grid.
Each side reads the wave and turns the result to bits. The team models that full chain with real gear terms. Strong Bell breaks in that chain would prove the code path works. Such breaks show non local links in a firm test. The work links code choice to test strength.
Catalog record
| CREATOR | Xiaotian Yang, Santiago Zamora, Rafael Chaves, Ulrik L. Andersen, Jonatan Bohr Brask, A. de Oliveira Junior |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-16189
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Robust Bell Nonlocality from Gottesman–Kitaev–Preskill States Xiaotian Yang, Santiago Zamora, Rafael Chaves, Ulrik L. Andersen, Jonatan Bohr Brask, A. de Oliveira Junior Abstract Bell tests based on homodyne detection are strongly constrained in continuous-variable systems. Can Gottesman–Kitaev–Preskill (GKP) encoding turn homodyne detection into a practical tool for revealing Bell nonlocality? We consider a physically motivated model in which each party performs homodyne detection and digitizes the continuous outcome via a fixed periodic binning, corresponding to logical Pauli measurements. Within this framework, we derive a bipartite no-go: CHSH cannot be violated for Bell-pair states. Moving beyond two parties, we show that finitely squeezed GKP-encoded GHZ and W states nevertheless exhibit strong multipartite nonlocality, violating multipartite Bell inequalities with homodyne-only readout. We quantify the required squeezing thresholds and robustness to loss, providing a route toward homodyne-based Bell tests in continuous-variable systems. Introduction– The violation of Bell inequalities decrees the death knell for local realism Einstein et al. 1935 ; Bell 1964 ; Bell 1966 ; Brunner et al. 2014 , while simultaneously providing an operational witness for certifying non-classicality in quantum technologies Ekert 1991 ; Acín et al. 2007 ; Pironio et al. 2010 ; Vazirani and Vidick 2014 . Making this witness experimentally meaningful, however, depends on the measurement scheme: if many events are lost due to detector inefficiencies, local-realistic models can reproduce the observed statistics, reopening the detection loophole Larsson 2014 . This raises the question of which measurements can reveal Bell nonlocality under realistic constraints. Homodyne detection is a natural candidate in this respect, as it can reach very high efficiencies and is therefore attractive for achieving loophole-free Bell tests García-Patrón et al. 2004a ; Eberhard 1993 . However, it also comes with a fundamental limitation because Bell nonlocality from Gaussian measurements on Gaussian states is impossible Bell and Aspect 2004 ; Johansen 1997 . Hence, non-Gaussian states must be used. While photon-subtracted states or optical cat states can circumvent this restriction Wenger et al. 2003 ; García-Patrón et al. 2004b ; García-Patrón et al. 2005 , their non-Gaussian features are fragile under loss, limiting their viability in realistic implementations. Such tests also require discretizing the continuous measurement outcomes into a finite set of results García-Patrón et al. 2004b ; García-Patrón et al. 2005 . This step is unavoidable when testing standard Bell inequalities, but it has nontrivial consequences: in many proposals the binning is optimized for a given state or measurement setting, which can significantly affect both the size and the robustness of the observed violation Wenger et al. 2003 ; Acín et al. 2009 ; Oudot et al. 2024 ; Lopetegui-González et al. 2026 . A fundamentally different and more robust route to address these issues is provided by the Gottesman–Kitaev–Preskill (GKP) encoding Gottesman et al. 2001 ; Glancy and Knill 2006 ; Grimsmo and Puri 2021 ; Tzitrin et al. 2020 . Figure 1: Bell nonlocality with Gottesman–Kitaev–Preskill States . Nonlocal correlations can be pulled out of continuous-variable systems using GKP states and only homodyne detection. Originally conceived for fault-tolerant quantum computing, GKP states encode qubits in one bosonic mode using a grid structure in phase space. This geometry leverages the oscillator’s infinite-dimensional Hilbert space to provide an inherent redundancy that would otherwise require hundreds of discrete physical qubits Noh and Chamberland 2020 . Crucially, the GKP framework allows for universal operations, error correction, and measurements using only Gaussian interactions Menicucci 2014 . Driven by the need for scalable quantum architectures, GKP states have been realised across diverse platforms, including trapped ions Flühmann et al. 2019 ; Hastrup et al. 2021 , microwave cavities Campagne-Ibarcq et al. 2020 , and integrated photonic sources Häussler and van Loock 2025 ; Larsen et al. 2025 . Figure 2: Ideal vs finite-energy GKP structure in phase space . Comparison between ideal and finite-energy (“realistic”) square GKP encodings in phase space and in the position quadrature. (a) sketch of the position-space marginal distribution showing that the ideal GKP code words form an infinitely sharp Dirac-comb lattice, while finite-energy GKP states replace each spike by a narrow Gaussian peak and suppress distant peaks under a broad Gaussian envelope (dashed curve), giving a normalizable state with finite mean energy. (b) Phase-space dot/contour plots of the Wigner functions associated with the logical GKP Pauli operators \mathbbm{1},X,Y and Z . Red (blue) marks positive (negative) regions, with the dot size indicating the relative magnitude of each lattice peak. Can the GKP encoding unlock robust Bell violations with homodyning? A first step toward bridging non-locality and the GKP code was taken via a device-independent quantum cryptography protocol Marshall and Weedbrook 2014 . Alice generates an entangled GKP state and shares one half with Bob to establish a secret key based on a CHSH violation. The drawback of this approach is that the violation requires a non-Gaussian T-gate, whose experimental implementation is challenging. This leads us to ask: given a GKP encoding, can we violate a Bell inequality using only Gaussian operations? Realistic GKP states are approximate, with finite squeezing and sensitivity to loss that can blur their grid structure. Whether this residual, noise-degraded non-Gaussianity is sufficient to overcome the homodyne no-go and yield an observable Bell violation has been unclear. In this work, we show that GKP states enable sizeable Bell violations using only homodyne detection, even in the presence of finite squeezing and noise (see Fig. 1 for a summary). Big picture– We characterise Bell nonlocality under a strict measurement constraint: each party performs homodyne detection of a quadrature, and the continuous outcome is converted into a binary result by a fixed binning rule. While the shared state may be non - Gaussian and highly entangled, this restriction yields a fundamental bipartite obstruction: Pauli measurements 1 1 1 Here “Pauli measurements” means measurements of the three Pauli observables X,Y,Z (not general qubit observables). on Bell states generate only local correlations in the CHSH scenario (see Appendix A ). This motivates our focus on multipartite Bell inequalities, for which Pauli measurements already lead to strong (and often optimal) violations. In quantum theory, Bell violations arise from entanglement probed by incompatible local measurements. We now introduce the two main ingredients of our continuous-variable implementation: the shared entangled resource and the local measurement settings. The resource is provided by embedding a logical qubit into a single bosonic mode using GKP grid states, while the measurement choices are implemented using homodyne detection followed by a fixed binary post-processing (binning) of the continuous outcome. Gottesman–Kitaev–Preskill as a resource– We use the square-lattice Gottesman–Kitaev–Preskill (GKP) encoding, which embeds a logical qubit into a single bosonic mode Gottesman et al. 2001 ; Grimsmo and Puri 2021 . In the \hat{q} -eigenbasis, the ideal (non-normalisable) logical codewords are Dirac-comb superpositions, \displaystyle\ket{0_{L}} \displaystyle\propto\sum_{s\in\mathbb{Z}}\ket{2s\sqrt{\pi}}_{q}, (1) \displaystyle\ket{1_{L}} \displaystyle\propto\sum_{s\in\mathbb{Z}}\ket{(2s+1)\sqrt{\pi}}_{q}, (2) where \ket{\xi}_{q} denotes the (generalised) eigenket of \hat{q} with eigenvalue \xi . These ideal states are unphysical, as they require infinite energy. To model realistic GKP states, we adopt the finite-energy approximation generated by the energy filter e^{-\epsilon\hat{n}} Matsuura et al. 2020 ; Royer et al. 2020 ; Hastrup and Andersen 2022 ; Hastrup and Andersen 2023 , where \hat{n}=\hat{a}^{\dagger}\hat{a}=\tfrac{1}{2}(\hat{q}^{2}+\hat{p}^{2}-1) . This regularisation replaces the Dirac comb by a lattice of narrow Gaussian peaks under a