aiqbench analysis
This research establishes rigorous lower bounds on the classical coordination cost required to simulate genuine global Kochen-Specker contextuality without invoking quantum resources. By demonstrating that simulating specific multipartite correlations requires significantly more classical bits than qubits—even under asymptotic repetition—the study quantifies a fundamental advantage of quantum information processing in causal online scenarios. The work advances our understanding of the resource trade-offs between classical communication and quantum entanglement in nonlocal systems.
Genuine Global Kochen-Specker Contextuality as Classical Coordination Cost
arXiv:2606.23577v3 Announce Type: replace Abstract: We formulate classical simulation of quantum correlations as coordination across a spacetime separator. All past-dependent information capable of influencing the future must cross the separator either through communicated messages or through a persistent classical state. We prove a general lower bound on this classical coordination cost in terms of the nonnegative rank of the separator correlation matrix, with approximate and amortized extensions governed by approximate nonnegative rank and Wyner common information. For sequential processes, we introduce a causal positive-realization rank that enforces consistent response and state-update maps, and we explain how bounded local computation and support-covering models arise as restricted cases. We apply the framework to genuine global Kochen--Specker contextuality, in which local subsystems are noncontextual and the tested multipartite blocks are generalized-Bell-local, while the complete empirical model has no global noncontextual explanation. At the symmetric point of a polarization--path Hardy construction, exact classical simulation requires two coordination bits, whereas the corresponding quantum separator requires one qubit. Under exact repetition, the classical lower bound is the base-two logarithm of three bits per copy; with asymptotically vanishing error, it becomes one and a half classical bits per copy, compared with one qubit per copy. We further construct a finite stabilizer--Peres--Mermin family for which the exact classical coordination cost grows quadratically with system size, while the quantum boundary cost grows only linearly. The separation applies to arbitrary finite-state causal online simulators. Establishing a constant-error quadratic separation and an intrinsic construction without flag conditioning remain open.
Abstract:We formulate classical simulation of quantum correlations as coordination across a spacetime separator. All past-dependent information capable of influencing the future must cross the separator either through communicated messages or through a persistent classical state. We prove a general lower bound on this classical coordination cost in terms of the nonnegative rank of the separator correlation matrix, with approximate and amortized extensions governed by approximate nonnegative rank and Wyner common information. For sequential processes, we introduce a causal positive-realization rank that enforces consistent response and state-update maps, and we explain how bounded local computation and support-covering models arise as restricted cases. We apply the framework to genuine global Kochen--Specker contextuality, in which local subsystems are noncontextual and the tested multipartite blocks are generalized-Bell-local, while the complete empirical model has no global noncontextual explanation. At the symmetric point of a polarization--path Hardy construction, exact classical simulation requires two coordination bits, whereas the corresponding quantum separator requires one qubit. Under exact repetition, the classical lower bound is the base-two logarithm of three bits per copy; with asymptotically vanishing error, it becomes one and a half classical bits per copy, compared with one qubit per copy. We further construct a finite stabilizer--Peres--Mermin family for which the exact classical coordination cost grows quadratically with system size, while the quantum boundary cost grows only linearly. The separation applies to arbitrary finite-state causal online simulators. Establishing a constant-error quadratic separation and an intrinsic construction without flag conditioning remain open.
44 pages. Added general nonnegative-rank coordination bounds, exact and asymptotic Hardy separations, an exact quadratic separation, and clarified the GLHV assumptions and support-covering framework. Comments and suggestions are welcome on alphaXiv
Quantum Physics (quant-ph); Computational Complexity (cs.CC)
arXiv:2606.23577 [quant-ph]
(or arXiv:2606.23577v3 [quant-ph] for this version)
Full text under the publisher's open licence · 288 words
Related coverage
Quantifiers and witnesses for the nonclassicality of measurements and of states
Quantum Journal · Jul 30, 2026
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · Sep 18, 2026
Global Kochen-Specker Contextuality Without Local Contextuality and Generalized Bell Nonlocality
arXiv · Jun 9, 2026
Finite Tests Rule Out Discrete Alternatives to Quantum Theory
Quantum Zeitgeist · Jul 19, 2026