Skip to content

[challenge]: Binary-answer monogamy-of-entanglement quantum advantage #254

Description

@kunyuan

Released by

Kun Chen, Institute of Theoretical Physics, Chinese Academy of Sciences

Contact email

chenkun0228@gmail.com

Method

Other

Challenge issue

Background

A standard monogamy-of-entanglement game asks two noncommunicating players to predict the outcome of a quantum measurement made by a referee. A finite game

$$G=(\Sigma,\Gamma,\pi,\{R_x^a\})$$

has a finite question set $\Sigma$, a finite answer set $\Gamma$, a question distribution $\pi$, and referee POVMs

$$R_x^a\succeq0,\qquad \sum_{a\in\Gamma}R_x^a=I_R.$$

The same sampled question $x$ is sent to both players. They win exactly when both return the referee's measurement outcome $a$.

Write $\omega(G)$ for the optimal unentangled value and $\omega^(G)$ for the optimal value when the players may share a finite-dimensional entangled strategy with the referee. Every two-question standard game satisfies $\omega(G)=\omega^(G)$. A four-question, three-answer separation is known. It remains open whether a separation is possible with the minimal nontrivial answer alphabet $|\Gamma|=2$.

Research objective

Construct one fully explicit finite standard monogamy-of-entanglement game with

$$|\Gamma|=2$$

and a legal finite-dimensional entangled strategy $S$ such that

$$p(G,S)>\omega(G).$$

Determining the exact global entangled optimum $\omega^(G)$ is unnecessary because $\omega^(G)\ge p(G,S)$. Since two-question games cannot separate the values, any positive witness necessarily has at least three questions.

The construction must remain a standard monogamy-of-entanglement game. A lossy or constrained extension, an ordinary Bell game, a communication-assisted protocol, or a coarse-grained statistic that does not define a legal binary referee POVM does not satisfy the challenge.

Success and verification gate

A complete result specifies exactly:

  • $\Sigma$, $\Gamma={0,1}$, and the probability distribution $\pi$;
  • the finite referee Hilbert space and every binary POVM ${R_x^0,R_x^1}$;
  • the shared state and both players' POVMs for the proposed strategy; and
  • exact values or rigorous certified intervals proving a strictly positive advantage.

An independent verifier should:

  1. check positivity, normalization, dimensions, and completeness of every state and POVM;
  2. compute the strategy's winning probability by exact or outward-rounded trace evaluation;
  3. recompute an exact or certified upper bound on $\omega(G)$—for finite data this can use deterministic-response enumeration and exact eigenvalue bounds, or an equivalent rigorous formulation; and
  4. confirm that the lower bound for $p(G,S)$ is strictly greater than the upper bound for $\omega(G)$.

The finite game and strategy are the answer. The solver's search process is not part of review.

Why this may lead to research output

A positive witness would locate quantum advantage at the smallest nontrivial answer alphabet and show that the growing family of binary no-advantage results is not universal. A universal negative theorem would instead prove that at least three answers are necessary. Monogamy-of-entanglement and extended nonlocal games also appear in quantum key distribution, relativistic bit commitment, and quantum position verification.

Current status and references

The binary-answer branch was audited on 28 July 2026 and assessed as likely open with medium confidence. The natural three-Pauli-basis binary game has no entangled advantage, and recent XOR/BB84 families provide additional no-advantage regimes, but no general binary impossibility theorem or positive witness was found.

  1. V. Russo, Extended Nonlocal Games.
  2. N. Johnston, R. Mittal, V. Russo, and J. Watrous, Extended non-local games and monogamy-of-entanglement games.
  3. L. Escolà-Farràs and F. Speelman, Lossy-and-Constrained Extended Non-Local Games with Applications to Quantum Cryptography.
  4. A. Broadbent and E. Culf, Rigidity for Monogamy-Of-Entanglement Games.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions