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:2609.24349

Robust self-testing of nonmaximal entanglement from a reduced inner product game

Ranyiliu Chen

quant-phSubmitted Sep 21, 2026• Announced Sep 22, 2026• 1 citation
arXivHTMLPDF

Abstract

We reduce Lalonde's pseudo-telepathy inner product game from six to five dimensions, keeping its four Alice questions and three Bob questions and reducing each answer alphabet to five. We prove that the reduced game self-tests its nonmaximally entangled state and all 35 measurement effects. The self-test is also robust: for a strategy winning with probability at least $1-\varepsilon$ we obtain $d_{\mathrm{ext}}\le\min\{1,2^{21}\sqrt{\varepsilon}\}$, where $d_{\mathrm{ext}}=\sqrt{1-Ξ}$ and $Ξ$ is the optimal squared fidelity with the reference state under local extraction channels. A compactness argument then gives simultaneous qualitative robustness for all measurement actions on the state under common local isometries. As a consequence of the self-test, no maximally entangled state of any finite dimension can win the game perfectly.

AI Summary

Can perfect winning in a nonlocal game certify a specific nonmaximally entangled state and all measurement actions? This paper studies a five-dimensional reduction of Lalonde’s inner product game, whose six-dimensional version already rules out perfect play with finite-dimensional maximal entanglement. The reduction retains four Alice questions and three Bob questions while reducing both answer alphabets from six to five. The reduced game has classical value $11/12$ and quantum value one. Every perfect finite-dimensional strategy self-tests the state $|\psi_\star\rangle=(2|00\rangle+2|11\rangle+2|22\rangle+2|33\rangle+|44\rangle)/\sqrt{17}$ and all 35 labeled measurement effects, up to local isometries and residual systems. Consequently, the perfect correlation table is unique, and no finite-dimensional maximally entangled state wins perfectly. For winning probability at least $1-\varepsilon$, local extraction achieves $d_{\mathrm{ext}}=\sqrt{1-\Xi}\leq\min\{1,2^{21}\sqrt{\varepsilon}\}$, where $\Xi$ is the optimal squared fidelity with the target under local channels. The results include mixed states and POVMs. The proof uses forbidden outcomes to recover five matching local labels, align four equal-amplitude components, and impose a fifth amplitude half as large through three binary agreement tests and one disagreement test. Pair sums and differences identify the measurements. Direct bounded-operator estimates yield the state bound; extending exact identities to commuting-operator representations enables a compactness argument for simultaneous, dimension-independent qualitative robustness of all measurement actions. The state prefactor is unoptimized and large, measurement robustness has no explicit rate, and no minimality claim is made for the dimension or question and answer counts.

Likely largely AI-generatedAI use disclosed
Several frontier LLMs was used under human direction to generate exploratory proofs, prepare and revise the manuscript.

Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.

Comments
10 + 21 pages

Related papers

The right entanglement for perfect nonlocality and zero-error communication

Laura Mančinska

We identify a structural mechanism by which perfect play in a nonlocal game requires and self-tests a specific non-maximally entangled state. To this end, we construct a Flagged Magic-Square game by extending the standard Magic-Square game with an additional ``flag'' outcome and two additional questions for each player. The game admits a perfect strategy using a state with Schmidt coefficients proportional to $(1,\sqrt{2},\sqrt{2},\sqrt{2},\sqrt{2})$. We give a direct algebraic proof that maximally entangled strategies have winning probability bounded away from one, independently of dimension. Moreover, the game self-tests this state, providing an example of a two-party pseudo-telepathy game that certifies a specific non-maximally entangled state. Our construction illustrates how adding suitable constraints to a rigid measurement geometry can enforce specific Schmidt ratios and provides a modular route to further self-tests based on other rigid pseudo-telepathy games. We establish a general game-to-channel reduction that transfers separations in the game model to one-shot zero-error communication over classical channels. Applied to the Flagged Magic-Square game, it yields a classical channel for which entanglement permits the zero-error transmission of ten messages in a single use, whereas any finite-dimensional maximally entangled resource permits at most nine. Thus, maximally entangled states are not a universally optimal entanglement resource for zero-error communication over classical noisy channels.

quant-pharXiv:2610.06512•Oct 6, 2026•1 citation

Maximally entangled states in pseudo-telepathy games

Laura Mančinska

A pseudo-telepathy game is a nonlocal game which can be won with probability one using some finite-dimensional quantum strategy but not using a classical one. Our central question is whether there exist two-party pseudo-telepathy games which cannot be won with probability one using a maximally entangled state. Towards answering this question, we develop conditions under which maximally entangled states suffice. In particular, we show that maximally entangled states suffice for weak projection games which we introduce as a relaxation of projection games. Our results also imply that any pseudo-telepathy weak projection game yields a device-independent certification of a maximally entangled state. In particular, by establishing connections to the setting of communication complexity, we exhibit a class of games $G_n$ for testing maximally entangled states of local dimension $Ω(n)$. We leave the robustness of these self-tests as an open question.

arXiv:1506.07080•Jun 23, 2015•5 citations

Maximally entangled states are not complete for pseudo-telepathy

Olivier Lalonde

One of the longstanding open problems in quantum nonlocality is to determine if maximally entangled states are complete for bipartite pseudo-telepathic games: namely, if every nonlocal game which admits a perfect entangled strategy admits such a strategy which uses a maximally entangled state. We exhibit a counterexample to this in the form of a bipartite nonlocal game with input sets of size 4 and 3 and output sets both of size 6. This game is part of a new class of nonlocal games, which we call inner product games, which could be of independent interest.

quant-pharXiv:2608.05378•Aug 7, 2026•2 citations

Minimum entangled state dimension required for pseudo-telepathy

Gilles Brassard, André Allan Méthot, Alain Tapp

Pseudo-telepathy provides an intuitive way of looking at Bell's inequalities, in which it is often obvious that feats achievable by use of quantum entanglement would be classically impossible. A~two-player pseudo-telepathy game proceeds as follows: Alice and Bob are individually asked a question and they must provide an answer. They are \emph{not} allowed any form of communication once the questions are asked, but they may have agreed on a common strategy prior to the execution of the game. We~say that they \emph{win} the game if the questions and answers fulfil a specific relation. A~game exhibits \emph{pseudo-telepathy} if there is a quantum strategy that makes Alice and Bob win the game for all possible questions, provided they share prior entanglement, whereas it would be impossible to win this game systematically in a classical setting. In~this paper, we show that any two-player pseudo-telepathy game requires the quantum players to share an entangled quantum system of dimension at least~\mbox{$3 \times 3$}. This is optimal for two-player games, but the most efficient pseudo-telepathy game possible, in terms of total dimension, involves \emph{three} players who share a quantum system of dimension~\mbox{$2 \times 2 \times 2$}.

arXiv:quant-ph/0412136•Quantum Information and Computation•Jul 1, 2005•36 citations

Glued magic games self-test maximally entangled states

Laura Mančinska, Thor Gabelgaard Nielsen, Jitendra Prakash

Self-testing results allow us to infer the underlying quantum mechanical description of states and measurements from classical outputs produced by non-communicating parties. The standard definition of self-testing does not apply in situations when there are two or more inequivalent optimal strategies. To address this, we introduce the notion of self-testing convex combinations of reference strategies, which is a generalisation of self-testing to multiple strategies. We show that the Glued Magic Square game [Quantum 4 (2020), p. 346] self-tests a convex combination of two inequivalent strategies. As a corollary, we obtain that the Glued Magic square game self-tests two EPR pairs thus answering an open question from [Quantum 4 (2020), p. 346]. Our self-test is robust and extends to natural generalisations of the Glued Magic Square game.

arXiv:2105.10658•May 22, 2021•9 citations

A Separation between Full-Rank PVM and Assumption-free Self-Testing

Ranyiliu Chen

We construct a nonlocal game that self-tests a maximally entangled qubit strategy among pure full-Schmidt-rank projective strategies, but admits an inequivalent optimum using one nonprojective measurement. This resolves a conjecture of Baptista et al. on imposing full rank and projectivity simultaneously. The construction combines CHSH with an auxiliary game $G$ that forces deterministic answers under these assumptions and admits a trine POVM optimum without them. We classify the optimal correlations of $G$ as a line segment parametrized by the tracial states of $\mathbb{C}\oplus M_2(\mathbb{C})$. Projectivity on the state support removes the matrix summand, whereas projective dilations with nonzero nonabort probability have a nontracial local state whose zero left ideal is not two-sided. Finally, a family in local dimension six shows that the full-rank PVM self-test is not robust.

quant-pharXiv:2609.10013•Sep 10, 2026•0 citations
We reduce Lalonde's pseudo-telepathy inner product game from six to five dimensions, keeping its four Alice questions and three Bob questions and reducing each answer alphabet to five. We prove that the reduced game self-tests its nonmaximally entangled state and all 35 measurement effects. The self-test is also robust: for a strategy winning with probability at least 1−ε1-\varepsilon1−ε we obtain dext≤min⁡{1,221ε}d_{\mathrm{ext}}\le\min\{1,2^{21}\sqrt{\varepsilon}\}dext​≤min{1,221ε​}, where dext=1−Ξd_{\mathrm{ext}}=\sqrt{1-Ξ}dext​=1−Ξ​ and ΞΞΞ is the optimal squared fidelity with the reference state under local extraction channels. A compactness argument then gives simultaneous qualitative robustness for all measurement actions on the state under common local isometries. As a consequence of the self-test, no maximally entangled state of any finite dimension can win the game perfectly.
Can perfect winning in a nonlocal game certify a specific nonmaximally entangled state and all measurement actions? This paper studies a five-dimensional reduction of Lalonde’s inner product game, whose six-dimensional version already rules out perfect play with finite-dimensional maximal entanglement. The reduction retains four Alice questions and three Bob questions while reducing both answer alphabets from six to five.
The reduced game has classical value 11/1211/1211/12 and quantum value one. Every perfect finite-dimensional strategy self-tests the state ∣ψ⋆⟩=(2∣00⟩+2∣11⟩+2∣22⟩+2∣33⟩+∣44⟩)/17|\psi_\star\rangle=(2|00\rangle+2|11\rangle+2|22\rangle+2|33\rangle+|44\rangle)/\sqrt{17}∣ψ⋆​⟩=(2∣00⟩+2∣11⟩+2∣22⟩+2∣33⟩+∣44⟩)/17​ and all 35 labeled measurement effects, up to local isometries and residual systems. Consequently, the perfect correlation table is unique, and no finite-dimensional maximally entangled state wins perfectly. For winning probability at least 1−ε1-\varepsilon1−ε, local extraction achieves dext=1−Ξ≤min⁡{1,221ε}d_{\mathrm{ext}}=\sqrt{1-\Xi}\leq\min\{1,2^{21}\sqrt{\varepsilon}\}dext​=1−Ξ​≤min{1,221ε​}, where Ξ\XiΞ is the optimal squared fidelity with the target under local channels. The results include mixed states and POVMs.
The proof uses forbidden outcomes to recover five matching local labels, align four equal-amplitude components, and impose a fifth amplitude half as large through three binary agreement tests and one disagreement test. Pair sums and differences identify the measurements. Direct bounded-operator estimates yield the state bound; extending exact identities to commuting-operator representations enables a compactness argument for simultaneous, dimension-independent qualitative robustness of all measurement actions.
The state prefactor is unoptimized and large, measurement robustness has no explicit rate, and no minimality claim is made for the dimension or question and answer counts.
We identify a structural mechanism by which perfect play in a nonlocal game requires and self-tests a specific non-maximally entangled state. To this end, we construct a Flagged Magic-Square game by extending the standard Magic-Square game with an additional ``flag'' outcome and two additional questions for each player. The game admits a perfect strategy using a state with Schmidt coefficients proportional to (1,2,2,2,2)(1,\sqrt{2},\sqrt{2},\sqrt{2},\sqrt{2})(1,2​,2​,2​,2​). We give a direct algebraic proof that maximally entangled strategies have winning probability bounded away from one, independently of dimension. Moreover, the game self-tests this state, providing an example of a two-party pseudo-telepathy game that certifies a specific non-maximally entangled state. Our construction illustrates how adding suitable constraints to a rigid measurement geometry can enforce specific Schmidt ratios and provides a modular route to further self-tests based on other rigid pseudo-telepathy games. We establish a general game-to-channel reduction that transfers separations in the game model to one-shot zero-error communication over classical channels. Applied to the Flagged Magic-Square game, it yields a classical channel for which entanglement permits the zero-error transmission of ten messages in a single use, whereas any finite-dimensional maximally entangled resource permits at most nine. Thus, maximally entangled states are not a universally optimal entanglement resource for zero-error communication over classical noisy channels.
A pseudo-telepathy game is a nonlocal game which can be won with probability one using some finite-dimensional quantum strategy but not using a classical one. Our central question is whether there exist two-party pseudo-telepathy games which cannot be won with probability one using a maximally entangled state. Towards answering this question, we develop conditions under which maximally entangled states suffice. In particular, we show that maximally entangled states suffice for weak projection games which we introduce as a relaxation of projection games. Our results also imply that any pseudo-telepathy weak projection game yields a device-independent certification of a maximally entangled state. In particular, by establishing connections to the setting of communication complexity, we exhibit a class of games GnG_nGn​ for testing maximally entangled states of local dimension Ω(n)Ω(n)Ω(n). We leave the robustness of these self-tests as an open question.
One of the longstanding open problems in quantum nonlocality is to determine if maximally entangled states are complete for bipartite pseudo-telepathic games: namely, if every nonlocal game which admits a perfect entangled strategy admits such a strategy which uses a maximally entangled state. We exhibit a counterexample to this in the form of a bipartite nonlocal game with input sets of size 4 and 3 and output sets both of size 6. This game is part of a new class of nonlocal games, which we call inner product games, which could be of independent interest.
Pseudo-telepathy provides an intuitive way of looking at Bell's inequalities, in which it is often obvious that feats achievable by use of quantum entanglement would be classically impossible. A~two-player pseudo-telepathy game proceeds as follows: Alice and Bob are individually asked a question and they must provide an answer. They are \emph{not} allowed any form of communication once the questions are asked, but they may have agreed on a common strategy prior to the execution of the game. We~say that they \emph{win} the game if the questions and answers fulfil a specific relation. A~game exhibits \emph{pseudo-telepathy} if there is a quantum strategy that makes Alice and Bob win the game for all possible questions, provided they share prior entanglement, whereas it would be impossible to win this game systematically in a classical setting. In~this paper, we show that any two-player pseudo-telepathy game requires the quantum players to share an entangled quantum system of dimension at least~\mbox{3×33 \times 33×3}. This is optimal for two-player games, but the most efficient pseudo-telepathy game possible, in terms of total dimension, involves \emph{three} players who share a quantum system of dimension~\mbox{2×2×22 \times 2 \times 22×2×2}.
Self-testing results allow us to infer the underlying quantum mechanical description of states and measurements from classical outputs produced by non-communicating parties. The standard definition of self-testing does not apply in situations when there are two or more inequivalent optimal strategies. To address this, we introduce the notion of self-testing convex combinations of reference strategies, which is a generalisation of self-testing to multiple strategies. We show that the Glued Magic Square game [Quantum 4 (2020), p. 346] self-tests a convex combination of two inequivalent strategies. As a corollary, we obtain that the Glued Magic square game self-tests two EPR pairs thus answering an open question from [Quantum 4 (2020), p. 346]. Our self-test is robust and extends to natural generalisations of the Glued Magic Square game.
We construct a nonlocal game that self-tests a maximally entangled qubit strategy among pure full-Schmidt-rank projective strategies, but admits an inequivalent optimum using one nonprojective measurement. This resolves a conjecture of Baptista et al. on imposing full rank and projectivity simultaneously. The construction combines CHSH with an auxiliary game GGG that forces deterministic answers under these assumptions and admits a trine POVM optimum without them. We classify the optimal correlations of GGG as a line segment parametrized by the tracial states of C⊕M2(C)\mathbb{C}\oplus M_2(\mathbb{C})C⊕M2​(C). Projectivity on the state support removes the matrix summand, whereas projective dilations with nonzero nonabort probability have a nontracial local state whose zero left ideal is not two-sided. Finally, a family in local dimension six shows that the full-rank PVM self-test is not robust.