Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2409.15203 · ACM Symposium on Theory of Computing (STOC)

Locality vs Quantum Codes

Samuel Dai, Ray Li

Published Jun 15, 2025• 9 citations
arXivHTMLPDF

Abstract

This paper proves optimal tradeoffs between the locality and parameters of quantum error-correcting codes. Quantum codes give a promising avenue towards quantum fault tolerance, but the practical constraint of locality limits their quality. The seminal Bravyi-Poulin-Terhal (BPT) bound says that a $[[n,k,d]]$ quantum stabilizer code with 2D-locality must satisfy $kd^2\le O(n)$. We answer the natural question: for better code parameters, how much "non-locality" is needed? In particular, (i) how long must the long-range interactions be, and (ii) how many long-range interactions must there be? We give a complete answer to both questions for all $n,k,d$: above the BPT bound, any 2D-embedding must have at least $Ω(\#^*)$ interactions of length $Ω(\ell^*)$, where $\#^*= \max(k,d)$ and $\ell^*=\max\big(\frac{d}{\sqrt{n}}, \big( \frac{kd^2}{n} \big)^{1/4} \big)$. Conversely, we exhibit quantum codes that show, in strong ways, that our interaction length $\ell^*$ and interaction count $\#^*$ are asymptotically optimal for all $n,k,d$. Our results generalize or improve all prior works on this question, including the BPT bound and the results of Baspin and Krishna. One takeaway of our work is that, for any desired distance $d$ and dimension $k$, the number of long-range interactions is asymptotically minimized by a good qLDPC code of length $Θ(\max(k,d))$. Following Baspin and Krishna, we also apply our results to the codes implemented in the stacked architecture and obtain better bounds. In particular, we rule out any implementation of hypergraph product codes in the stacked architecture.

DOI
10.1145/3717823.3718113

Related papers

Improved rate-distance trade-offs for quantum codes with restricted connectivity

Nouédyn Baspin, Venkatesan Guruswami, Anirudh Krishna +1 more

Abstract For quantum error-correcting codes to be realizable, it is important that the qubits subject to the code constraints exhibit some form of limited connectivity. The works of Bravyi and Terhal (2009 New J. Phys. 11 043029) (BT) and Bravyi et al (2010 Phys. Rev. Lett. 104 050503) (BPT) established that geometric locality constrains code properties—for instance [ [ n , k , d ] ] quantum codes defined by local checks on the D-dimensional lattice must obey k d 2 / ( D − 1 ) ⩽ O ( n ) . Baspin and Krishna (2022 Quantum 6 711) studied the more general question of how the connectivity graph associated with a quantum code constrains the code parameters. These trade-offs apply to a richer class of codes compared to the BPT and BT bounds, which only capture geometrically-local codes. We extend and improve this work, establishing a tighter dimension-distance trade-off as a function of the size of separators in the connectivity graph. We also obtain a distance bound that covers all stabilizer codes with a particular separation profile, rather than only LDPC codes.

arXiv:2307.03283•Quantum Science and Technology•Oct 4, 2024•17 citations

Intrinsic locality dimension of quantum codes

Yimin Lu, Esther Xiaozhen Fu, Zi-Wen Liu

Quantum error-correcting codes are a cornerstone of quantum computing, with broad and profound connections to physics and mathematics. In this work, we introduce the notion of intrinsic locality dimension of stabilizer codes, which is independent of the underlying geometry of quantum codes and naturally extends to non-integer values. Drawing on mathematical tools from fractal geometry and geometric measure theory, the intrinsic locality dimension accommodates flexible architectures and provides a quantitative measure of code connectivity, encompassing both topological codes and algebraic constructions such as bivariate-bicycle-type codes. We show how the intrinsic dimension serves as a fundamental organizing parameter that unifies code properties. In particular, we prove general limitations on code parameters and compatible fault-tolerant logical gates induced by the intrinsic dimension, generalizing the Bravyi--Poulin--Terhal and Bravyi--König bounds for regular topological codes, respectively. Furthermore, we consider implications on thermal properties: toward fully characterizing the geometry requirement for self-correcting quantum memories (SCQMs), we present a conditional no-go result for SCQMs in dimension $3-ε$ and take stock of existing results on low-dimensional SCQMs. Our theory provides a unifying mathematical framework for understanding the fundamental capabilities and geometric implementations of quantum error correction and fault tolerance.

quant-pharXiv:2605.31441•Jun 1, 2026•3 citations

Quantifying Nonlocality: How Outperforming Local Quantum Codes Is Expensive

Nouédyn Baspin, Anirudh Krishna

Quantum low-density parity-check (LDPC) codes are a promising avenue to reduce the cost of constructing scalable quantum circuits. However, it is unclear how to implement these codes in practice. Seminal results of Bravyi et al. [Phys. Rev. Lett. 104, 050503 (2010)] have shown that quantum LDPC codes implemented through local interactions obey restrictions on their dimension $k$ and distance $d$. Here we address the complementary question of how many long-range interactions are required to implement a quantum LDPC code with parameters $k$ and $d$. In particular, in 2D we show that a quantum LDPC code with distance $d\ensuremath{\propto}{n}^{1/2+ϵ}$ requires $\mathrm{\ensuremath{\Omega}}({n}^{1/2+ϵ})$ interactions of length $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}({n}^{ϵ})$. Further, a code satisfying $k\ensuremath{\propto}n$ with distance $d\ensuremath{\propto}{n}^{\ensuremath{\alpha}}$ requires $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}(n)$ interactions of length $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}({n}^{\ensuremath{\alpha}/2})$. As an application of these results, we consider a model called a stacked architecture, which has previously been considered as a potential way to implement quantum LDPC codes. In this model, although most interactions are local, a few of them are allowed to be very long. We prove that limited long-range connectivity implies quantitative bounds on the distance and code dimension.

arXiv:2109.10982•Physical Review Letters•Jul 28, 2022•42 citations

The locality of quantum codes

Nouédyn Baspin

This thesis concerns itself with the role locality plays in the performance of quantum codes. We first address the difficulty of producing codestates from relatively good codes, specifically we investigate the depth of geometrically local encoding circuits. Previous results struggled to produce bounds allowing for general, non-local classical communications in the circuit. We show that, owed to the entanglement hidden in codestates, the depth of encoding circuits is lower bounded, even when allowing for boundless classical computation. We then move on to address a converse question: given a particular code, what can we say about its geometrical structure? We show that implementing stabiliser codes requires satisfying some connectivity requirements that depend on k and d. We quantify these requirements through the notion of graph expansion. Finally, in the last chapter, we ask the question of the achievability of these requirements. The celebrated BPT bound states upper bound on the parameters that can be achieved for codes local in three dimensions. We provide the first construction that matches this bound.

•Jan 1, 2025•0 citations

Optimal Locality and Parameter Tradeoffs for Subsystem Codes

Dai, Samuel, Ray Li, Tang, Eugene

We study the tradeoffs between the locality and parameters of subsystem codes. We prove lower bounds on both the number and lengths of interactions in any D-dimensional embedding of a subsystem code. Specifically, we show that any embedding of a subsystem code with parameters [[n,k,d]] into R^D must have at least M^* interactions of length at least 𝓁^*, where M^* = Ω(max(k,d)), and 𝓁^* = Ω(max(d/(n^((D-1)/D)), ((kd^(1/(D-1))/n))^((D-1)/D))). We also give tradeoffs between the locality and parameters of commuting projector codes in D-dimensions, generalizing a result of Dai and Li [Dai and Li, 2025]. We provide explicit constructions of embedded codes that show our bounds are optimal in both the interaction count and interaction length.

arXiv:2503.22651•Leibniz international proceedings in informatics•Jan 1, 2025•1 citation

Fault-tolerant logical gates in quantum error-correcting codes

Fernando Pastawski, Beni Yoshida

Recently, S. Bravyi and R. K\"onig [Phys. Rev. Lett. 110, 170503 (2013)] have shown that there is a trade-off between fault-tolerantly implementable logical gates and geometric locality of stabilizer codes. They consider locality-preserving operations which are implemented by a constant-depth geometrically local circuit and are thus fault tolerant by construction. In particular, they show that, for local stabilizer codes in $D$ spatial dimensions, locality-preserving gates are restricted to a set of unitary gates known as the $D\mathrm{th}$ level of the Clifford hierarchy. In this paper, we explore this idea further by providing several extensions and applications of their characterization to qubit stabilizer and subsystem codes. First, we present a no-go theorem for self-correcting quantum memory. Namely, we prove that a three-dimensional stabilizer Hamiltonian with a locality-preserving implementation of a non-Clifford gate cannot have a macroscopic energy barrier. This result implies that non-Clifford gates do not admit such implementations in Haah's cubic code and Michnicki's welded code. Second, we prove that the code distance of a $D$-dimensional local stabilizer code with a nontrivial locality-preserving $m\mathrm{th}$-level Clifford logical gate is upper bounded by $O({L}^{D+1\ensuremath{-}m})$. For codes with non-Clifford gates ($m>2$), this improves the previous best bound by S. Bravyi and B. Terhal [New. J. Phys. 11, 043029 (2009)]. Topological color codes, introduced by H. Bombin and M. A. Martin-Delgado [Phys. Rev. Lett. 97, 180501 (2006); Phys. Rev. Lett. 98, 160502 (2007); Phys. Rev. B 75, 075103 (2007)], saturate the bound for $m=D$. Third, we prove that the qubit erasure threshold for codes with a nontrivial transversal $m\mathrm{th}$-level Clifford logical gate is upper bounded by $1/m$. This implies that no family of fault-tolerant codes with transversal gates in increasing level of the Clifford hierarchy may exist. This result applies to arbitrary stabilizer and subsystem codes and is not restricted to geometrically local codes. Fourth, we extend the result of Bravyi and K\"onig to subsystem codes. Unlike stabilizer codes, the so-called union lemma does not apply to subsystem codes. This problem is avoided by assuming the presence of an error threshold in a subsystem code, and a conclusion analogous to that of Bravyi and K\"onig is recovered.

arXiv:1408.1720•Physical Review A•Jan 8, 2015•110 citations
This paper proves optimal tradeoffs between the locality and parameters of quantum error-correcting codes. Quantum codes give a promising avenue towards quantum fault tolerance, but the practical constraint of locality limits their quality. The seminal Bravyi-Poulin-Terhal (BPT) bound says that a [[n,k,d]][[n,k,d]][[n,k,d]] quantum stabilizer code with 2D-locality must satisfy kd2≤O(n)kd^2\le O(n)kd2≤O(n). We answer the natural question: for better code parameters, how much "non-locality" is needed? In particular, (i) how long must the long-range interactions be, and (ii) how many long-range interactions must there be? We give a complete answer to both questions for all n,k,dn,k,dn,k,d: above the BPT bound, any 2D-embedding must have at least Ω(#∗)Ω(\#^*)Ω(#∗) interactions of length Ω(ℓ∗)Ω(\ell^*)Ω(ℓ∗), where #∗=max⁡(k,d)\#^*= \max(k,d)#∗=max(k,d) and ℓ∗=max⁡(dn,(kd2n)1/4)\ell^*=\max\big(\frac{d}{\sqrt{n}}, \big( \frac{kd^2}{n} \big)^{1/4} \big)ℓ∗=max(n​d​,(nkd2​)1/4). Conversely, we exhibit quantum codes that show, in strong ways, that our interaction length ℓ∗\ell^*ℓ∗ and interaction count #∗\#^*#∗ are asymptotically optimal for all n,k,dn,k,dn,k,d. Our results generalize or improve all prior works on this question, including the BPT bound and the results of Baspin and Krishna. One takeaway of our work is that, for any desired distance ddd and dimension kkk, the number of long-range interactions is asymptotically minimized by a good qLDPC code of length Θ(max⁡(k,d))Θ(\max(k,d))Θ(max(k,d)). Following Baspin and Krishna, we also apply our results to the codes implemented in the stacked architecture and obtain better bounds. In particular, we rule out any implementation of hypergraph product codes in the stacked architecture.
Abstract For quantum error-correcting codes to be realizable, it is important that the qubits subject to the code constraints exhibit some form of limited connectivity. The works of Bravyi and Terhal (2009 New J. Phys. 11 043029) (BT) and Bravyi et al (2010 Phys. Rev. Lett. 104 050503) (BPT) established that geometric locality constrains code properties—for instance [ [ n , k , d ] ] quantum codes defined by local checks on the D-dimensional lattice must obey k d 2 / ( D − 1 ) ⩽ O ( n ) . Baspin and Krishna (2022 Quantum 6 711) studied the more general question of how the connectivity graph associated with a quantum code constrains the code parameters. These trade-offs apply to a richer class of codes compared to the BPT and BT bounds, which only capture geometrically-local codes. We extend and improve this work, establishing a tighter dimension-distance trade-off as a function of the size of separators in the connectivity graph. We also obtain a distance bound that covers all stabilizer codes with a particular separation profile, rather than only LDPC codes.
Quantum error-correcting codes are a cornerstone of quantum computing, with broad and profound connections to physics and mathematics. In this work, we introduce the notion of intrinsic locality dimension of stabilizer codes, which is independent of the underlying geometry of quantum codes and naturally extends to non-integer values. Drawing on mathematical tools from fractal geometry and geometric measure theory, the intrinsic locality dimension accommodates flexible architectures and provides a quantitative measure of code connectivity, encompassing both topological codes and algebraic constructions such as bivariate-bicycle-type codes. We show how the intrinsic dimension serves as a fundamental organizing parameter that unifies code properties. In particular, we prove general limitations on code parameters and compatible fault-tolerant logical gates induced by the intrinsic dimension, generalizing the Bravyi--Poulin--Terhal and Bravyi--König bounds for regular topological codes, respectively. Furthermore, we consider implications on thermal properties: toward fully characterizing the geometry requirement for self-correcting quantum memories (SCQMs), we present a conditional no-go result for SCQMs in dimension 3−ε3-ε3−ε and take stock of existing results on low-dimensional SCQMs. Our theory provides a unifying mathematical framework for understanding the fundamental capabilities and geometric implementations of quantum error correction and fault tolerance.
Quantum low-density parity-check (LDPC) codes are a promising avenue to reduce the cost of constructing scalable quantum circuits. However, it is unclear how to implement these codes in practice. Seminal results of Bravyi et al. [Phys. Rev. Lett. 104, 050503 (2010)] have shown that quantum LDPC codes implemented through local interactions obey restrictions on their dimension kkk and distance ddd. Here we address the complementary question of how many long-range interactions are required to implement a quantum LDPC code with parameters kkk and ddd. In particular, in 2D we show that a quantum LDPC code with distance $d\ensuremath{\propto}{n}^{1/2+ϵ}$ requires $\mathrm{\ensuremath{\Omega}}({n}^{1/2+ϵ})$ interactions of length $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}({n}^{ϵ})$. Further, a code satisfying $k\ensuremath{\propto}n$ with distance $d\ensuremath{\propto}{n}^{\ensuremath{\alpha}}$ requires $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}(n)$ interactions of length $\stackrel{\texttildelow{}}{\mathrm{\ensuremath{\Omega}}}({n}^{\ensuremath{\alpha}/2})$. As an application of these results, we consider a model called a stacked architecture, which has previously been considered as a potential way to implement quantum LDPC codes. In this model, although most interactions are local, a few of them are allowed to be very long. We prove that limited long-range connectivity implies quantitative bounds on the distance and code dimension.
This thesis concerns itself with the role locality plays in the performance of quantum codes. We first address the difficulty of producing codestates from relatively good codes, specifically we investigate the depth of geometrically local encoding circuits. Previous results struggled to produce bounds allowing for general, non-local classical communications in the circuit. We show that, owed to the entanglement hidden in codestates, the depth of encoding circuits is lower bounded, even when allowing for boundless classical computation. We then move on to address a converse question: given a particular code, what can we say about its geometrical structure? We show that implementing stabiliser codes requires satisfying some connectivity requirements that depend on k and d. We quantify these requirements through the notion of graph expansion. Finally, in the last chapter, we ask the question of the achievability of these requirements. The celebrated BPT bound states upper bound on the parameters that can be achieved for codes local in three dimensions. We provide the first construction that matches this bound.
We study the tradeoffs between the locality and parameters of subsystem codes. We prove lower bounds on both the number and lengths of interactions in any D-dimensional embedding of a subsystem code. Specifically, we show that any embedding of a subsystem code with parameters [[n,k,d]] into R^D must have at least M^* interactions of length at least 𝓁^*, where M^* = Ω(max(k,d)), and 𝓁^* = Ω(max(d/(n^((D-1)/D)), ((kd^(1/(D-1))/n))^((D-1)/D))). We also give tradeoffs between the locality and parameters of commuting projector codes in D-dimensions, generalizing a result of Dai and Li [Dai and Li, 2025]. We provide explicit constructions of embedded codes that show our bounds are optimal in both the interaction count and interaction length.
Recently, S. Bravyi and R. K\"onig [Phys. Rev. Lett. 110, 170503 (2013)] have shown that there is a trade-off between fault-tolerantly implementable logical gates and geometric locality of stabilizer codes. They consider locality-preserving operations which are implemented by a constant-depth geometrically local circuit and are thus fault tolerant by construction. In particular, they show that, for local stabilizer codes in DDD spatial dimensions, locality-preserving gates are restricted to a set of unitary gates known as the DthD\mathrm{th}Dth level of the Clifford hierarchy. In this paper, we explore this idea further by providing several extensions and applications of their characterization to qubit stabilizer and subsystem codes. First, we present a no-go theorem for self-correcting quantum memory. Namely, we prove that a three-dimensional stabilizer Hamiltonian with a locality-preserving implementation of a non-Clifford gate cannot have a macroscopic energy barrier. This result implies that non-Clifford gates do not admit such implementations in Haah's cubic code and Michnicki's welded code. Second, we prove that the code distance of a DDD-dimensional local stabilizer code with a nontrivial locality-preserving mthm\mathrm{th}mth-level Clifford logical gate is upper bounded by $O({L}^{D+1\ensuremath{-}m})$. For codes with non-Clifford gates (m>2m>2m>2), this improves the previous best bound by S. Bravyi and B. Terhal [New. J. Phys. 11, 043029 (2009)]. Topological color codes, introduced by H. Bombin and M. A. Martin-Delgado [Phys. Rev. Lett. 97, 180501 (2006); Phys. Rev. Lett. 98, 160502 (2007); Phys. Rev. B 75, 075103 (2007)], saturate the bound for m=Dm=Dm=D. Third, we prove that the qubit erasure threshold for codes with a nontrivial transversal mthm\mathrm{th}mth-level Clifford logical gate is upper bounded by 1/m1/m1/m. This implies that no family of fault-tolerant codes with transversal gates in increasing level of the Clifford hierarchy may exist. This result applies to arbitrary stabilizer and subsystem codes and is not restricted to geometrically local codes. Fourth, we extend the result of Bravyi and K\"onig to subsystem codes. Unlike stabilizer codes, the so-called union lemma does not apply to subsystem codes. This problem is avoided by assuming the presence of an error threshold in a subsystem code, and a conclusion analogous to that of Bravyi and K\"onig is recovered.