Quantum Nonlocality under Latency Constraints

Dawei Ding, Zhengfeng Ji, Pierre Pocreau, Mingze Xu, Xin-Yu XuarXiv 2025 · DOI · arXiv

Abstract

Bell inequality violation is the phenomenon where multiple non-communicating parties can exhibit correlations using quantum resources that are impossible if they can only use classical resources. One way to enforce non-communication is to apply a latency constraint: the parties must all produce outputs after they receive their inputs within a time window shorter than the speed of light delay between any pair of parties. If this latency constraint is relaxed so that a subset of the parties can communicate, we can obtain a new set of inequalities on correlations that extends Bell inequalities in a very natural way. Moreover, with this relaxed latency constraint, we can also have quantum communication between a subset of parties and thereby achieve possible quantum violations of these new inequalities. We ultimately wish to answer the fundamental question: "What are the physically realizable correlations between multiple parties under varying latency constraints?" To answer this question, we introduce latency-constrained games, a mathematical framework that extends nonlocal games to the setting where a subset of parties can communicate. The notion of latency-constrained games can have real-world applications, including high frequency trading, distributed computing, computer architecture, and distributed control systems.

Introduction

In quantum foundations, Bell nonlocality is traditionally viewed as a binary state: either parties are perfectly isolated, leading to potential Bell inequality violations, or they can communicate freely, rendering the situation trivial. However, real-world systems operate in the "in-between." Communication is limited by physical distance and the speed of light. This paper introduces a framework to explore the spectrum of correlations possible when communication is partially restricted by latency.

By relaxing the strict "no-communication" rule of standard Bell tests, the authors identify new fundamental bounds on classical correlations. They demonstrate how quantum mechanics can surpass these bounds, providing a measurable advantage in time-sensitive tasks such as high-frequency trading or distributed control systems.

The Framework: Latency-Constrained Games

The authors extend the standard "nonlocal game" framework into two distinct categories:

  1. Simple LC Games: Parties are connected by a directed graph $G$. An edge from Alice to Bob indicates that Alice can send information to Bob within a single round before outputs are required.
  2. Multi-Step LC Games: A more complex model involving multiple rounds of inputs and outputs over time, governed by a latency function $\ell(i, j)$ that defines the time steps required for information to travel between parties.
Space-time diagrams illustrating the hierarchy of communication scenarios (Bell, partial, back-and-forth, and full) that emerge as the latency constraint $t$ is relaxed.

As shown in Figure 2, varying the time window ($t$) allows for different levels of interaction. At $t=0$, we have the standard Bell scenario. As $t$ increases, parties can begin to communicate in one direction, then back-and-forth, until the scenario becomes fully communicating.

Technical Innovations

To analyze these games, the authors developed several key technical tools:

  • Forwarding Strategies: A class of classical strategies where parties simply "forward" their inputs to others. The paper proves that general quantum strategies are strictly more powerful than these due to the "monogamy of entanglement."
  • Generalized See-Saw Algorithm: To find maximum quantum winning probabilities, the authors adapted the "see-saw" optimization technique. They utilized Choi operators and the link product formalism from quantum comb theory to manage the sequential nature of communication and measurement.

Key Results: The Quantum Time Advantage

The core finding of the research is the "Quantum Time Advantage." Quantum players can achieve specific winning probabilities much faster than classical players under the same speed-of-light constraints.

  • The Distributed CHSH Miracle: In a scenario where two parties ($A$ and $B$) can communicate but a third ($C$) is distant, classical players are bounded by a winning probability of $0.75$. Quantum players, using only a single round of communication, can reach the full Tsirelson bound of $\cos^2(\pi/8) \approx 0.853$.
  • Numerical Gaps: In tests involving 50 random extended XOR games, a clear separation between classical "forwarding" and general quantum strategies was found in 70% of the cases.
  • Latency Step-Functions: When plotting game value against time, the results show a clear hierarchy: as the latency constraint increases, the quantum value jumps to higher plateaus earlier than the classical value.

Conclusion and Future Work

This work proves that communication does not "wash out" nonlocality; rather, it creates new "Bell-like" inequalities that quantum players can violate. While the current research provides robust lower bounds for quantum performance, future work aims to establish absolute upper bounds via a generalized NPA hierarchy. Additionally, the authors suggest extending the model to curved spacetime to explore how gravity might influence these latency-constrained correlations.

Summary updated

Abstract

Bell inequality violation is the phenomenon where multiple non-communicating parties can exhibit correlations using quantum resources that are impossible if they can only use classical resources. One way to enforce non-communication is to apply a latency constraint: the parties must all produce outputs after they receive their inputs within a time window shorter than the speed of light delay between any pair of parties. If this latency constraint is relaxed so that a subset of the parties can communicate, we can obtain a new set of inequalities on correlations that extends Bell inequalities in a very natural way. Moreover, with this relaxed latency constraint, we can also have quantum communication between a subset of parties and thereby achieve possible quantum violations of these new inequalities. We ultimately wish to answer the fundamental question: "What are the physically realizable correlations between multiple parties under varying latency constraints?" To answer this question, we introduce latency-constrained games, a mathematical framework that extends nonlocal games to the setting where a subset of parties can communicate. The notion of latency-constrained games can have real-world applications, including high frequency trading, distributed computing, computer architecture, and distributed control systems.
BibTeX
@misc{ding2025quantum,
  author = {Dawei Ding and Zhengfeng Ji and Pierre Pocreau and Mingze Xu and Xin-Yu Xu},
  title = {{Quantum Nonlocality under Latency Constraints}},
  year = {2025},
  eprint = {2510.26349},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2510.26349},
  howpublished = {arXiv},
}

Loading related content...

Figure