All pure entangled states can lead to fully nonlocal correlations

This paper demonstrates that maximal entanglement is not a prerequisite for fully nonlocal correlations by establishing a link to antidistinguishability to prove that all pure entangled states in dimensions $d \geq 3$ can exhibit full nonlocality, either directly through specific Schmidt coefficient conditions or via activation in a many-copy scenario.

The Gist

Imagine the universe has a set of secret rules that govern how particles interact. For a long time, physicists have known that quantum particles can be "entangled," meaning they share a connection so strong that measuring one instantly tells you about the other, no matter how far apart they are. This defies our everyday logic, which assumes that objects only influence their immediate surroundings. This phenomenon is called nonlocality.

However, not all nonlocal connections are created equal. Some are "weakly" nonlocal, meaning you could still explain them with a slightly tweaked version of classical logic (like a hidden script the particles are following). Others are fully nonlocal. These are the "superstars" of quantum weirdness. They are so strange that no amount of tweaking classical logic can ever explain them. It's like trying to explain a magic trick using only the laws of physics; it's simply impossible.

For decades, scientists thought that only the most perfectly balanced, "maximally entangled" states could achieve this "fully nonlocal" status. They believed that if the entanglement was even slightly imperfect (non-maximally entangled), the connection would be too "fuzzy" to break classical logic completely.

The Big Discovery
This paper shatters that belief. The authors prove that imperfectly entangled states can also be fully nonlocal, provided the particles are in a space of three dimensions or higher (like a 3D dice roll rather than a simple coin flip).

To do this, they built a bridge between two seemingly unrelated concepts:

1. The Magic Game: A scenario where two players (Alice and Bob) must coordinate answers to win a game that is impossible to win if they are just following a pre-written script (local hidden variables).
2. The "Impossible to Spot" Test: A concept called antidistinguishability. Imagine you have a bag of three different colored balls. "Distinguishability" means you can look at a ball and know exactly which color it is. "Antidistinguishability" is the reverse: you can look at a ball and be 100% certain it is not one of the other specific colors.

The Analogy: The Detective and the Suspects
Think of the quantum state as a set of suspects.

• Maximally entangled states are like a lineup of suspects who are all so distinct that a detective can instantly rule out three of them just by looking at the fourth.
• Non-maximally entangled states are like suspects who look very similar. The detective usually struggles to tell them apart.

The authors discovered a clever trick. Even if the suspects look similar (imperfect entanglement), if the detective asks the right specific questions (measurements), they can still rule out every single "scripted" possibility. They proved that for any level of imperfection, there is a way to set up the game so that the quantum players win with 100% certainty, while any classical team following a script must fail.

Key Findings in Plain English

1. Imperfect is Okay: You don't need a "perfect" quantum connection to break the laws of classical physics. As long as the particles are in a high enough dimension (3D or more), even a slightly "lopsided" connection can be fully nonlocal.
2. The Magic of Copying: What if you have a very weak, imperfect connection that can't break classical logic on its own? The paper shows that if you take multiple copies of that same weak connection and use them together, they can "activate" each other. It's like having a single weak flashlight that can't light up a room, but if you stack ten of them together, they suddenly become bright enough to banish the shadows. Any pure entangled state, no matter how weak, can be made fully nonlocal if you have enough copies.
3. The Limit: Not every state is a superstar. The authors also found that there are specific "weak" states that, even with all the tricks in the book, still have a tiny bit of "classical logic" hiding inside them. They can never be fully nonlocal on their own.

Why This Matters
The paper doesn't just say "we found a new state." It provides a simple checklist (based on the "size" of the entanglement) to tell you if a state is strong enough to be fully nonlocal. It also proves that the "perfect" state isn't the only key to the kingdom; the "imperfect" ones have their own superpowers if you know how to unlock them.

In short: The universe is even more flexible than we thought. You don't need perfection to achieve the impossible; sometimes, a little bit of imperfection, combined with the right strategy, is enough to break the rules of classical reality.

Technical Summary

1. Problem Statement

The paper addresses a fundamental open question in quantum foundations: Is maximal entanglement necessary for a quantum state to exhibit "full nonlocality"?

• Full Nonlocality: A quantum correlation is "fully nonlocal" if it cannot be decomposed into any local hidden variable (LHV) model, even with an arbitrarily small local component. Mathematically, the local content ($LC$) of the correlation is zero ($LC=0$). These correlations saturate the non-signaling bound of Bell inequalities (often called "pseudo-telepathy" or "all-versus-nothing" proofs).
• The Gap: It was previously known that:
  • Maximally entangled states (MES) in dimensions $d \geq 3$ exhibit full nonlocality.
  • Pure non-maximally entangled states (NMES) of two qubits ($d=2$) do not exhibit full nonlocality in the single-copy scenario.
  • It was unknown whether NMES in higher dimensions ($d \geq 3$) could exhibit full nonlocality, or if full nonlocality was exclusive to MES.

2. Methodology

The authors develop a novel theoretical framework linking full nonlocality to the concept of antidistinguishability of quantum states.

• Antidistinguishability: A set of states $\{\rho_i\}$ is antidistinguishable if there exists a measurement $\{M_i\}$ such that $\text{Tr}(M_i \rho_i) = 0$ for all $i$. This means the measurement outcome $i$ definitively excludes the possibility that the state was $\rho_i$.
• The Connection (Theorem 2): The authors prove that a bipartite state is fully nonlocal if and only if there exist measurements for Alice such that for every possible deterministic local strategy (outcome string $\alpha$), the resulting set of post-measurement states on Bob's side is antidistinguishable.
  • If the states are antidistinguishable, Bob can choose a measurement that yields a probability of 0 for the specific outcome combination required by Alice's deterministic strategy.
  • This forces the probability of that specific local strategy to be zero in the quantum distribution, effectively eliminating all local deterministic strategies from the decomposition, resulting in $LC=0$.
• Tools Used:
  • Semidefinite Programming (SDP): To analyze antidistinguishability conditions.
  • Mutually Unbiased Bases (MUBs): Used as measurement settings to derive sufficient conditions for antidistinguishability based on the overlap of post-measurement states.
  • Gram Matrix Analysis: Utilizing the Frobenius norm of the Gram matrix of post-measurement states to apply sufficient criteria for antidistinguishability (Theorem 1).

3. Key Contributions and Results

A. Existence of Fully Nonlocal Non-Maximally Entangled States

The authors prove that full nonlocality is not restricted to maximally entangled states.

• Theorem 3: They derive simple sufficient conditions based on the Schmidt coefficients ($\lambda_{max}$ and $\lambda_{min}$) of a pure entangled state $|\psi\rangle = \sum \sqrt{\lambda_i}|ii\rangle$.
• Result 1: In every Hilbert space dimension $d \geq 3$, there exist non-maximally entangled states that are fully nonlocal.
  • For $d=3$ (qutrits), they construct a specific set of 5 measurements (extending beyond standard MUBs) to prove the existence of fully nonlocal partially entangled states.
  • For $d \geq 5$, they show that states with specific bounds on $\lambda_{max}$ and $\lambda_{min}$ (e.g., $\lambda_{max} \leq 1/4$) are fully nonlocal.
  • Significance: They even construct examples where $\lambda_{max} > 1/2$. These states cannot be deterministically transformed into a maximally entangled state via Local Operations and Classical Communication (LOCC), proving that full nonlocality is an intrinsic property of certain non-maximally entangled states, not just a feature of MES.

B. Activation of Full Nonlocality

• Result 2: They prove that all pure entangled states (including non-maximally entangled qubit pairs) can be activated to show full nonlocality in a many-copy scenario.
• Mechanism: By sharing $k$ copies of a state $|\psi\rangle^{\otimes k}$ and performing separable measurements, the overlaps between post-measurement states decrease exponentially with $k$.
• Implication: For any pure entangled state, there exists a finite number of copies $k$ such that the resulting system is fully nonlocal. This generalizes previous results which were limited to specific cases.

C. Limits: States with Non-Zero Local Content

• Result 3 & 4: The authors also identify families of states that do not exhibit full nonlocality, even in the single-copy scenario with arbitrary measurements (POVMs).
• Condition: If the largest Schmidt coefficient $\lambda_{max}$ is sufficiently large (specifically $\lambda_{max} > \lambda_1(d-1)^2$ for projective measurements, or a stricter bound for general POVMs), the state retains a non-zero local content ($LC > 0$).
• Significance: This establishes that full nonlocality is not a universal property of all pure entangled states in the single-copy limit; there is a "gap" where states are entangled but not fully nonlocal.

4. Significance and Impact

1. Foundational Insight: The work resolves the open question of whether maximal entanglement is a prerequisite for full nonlocality. The answer is no; a broad class of non-maximally entangled states in dimensions $d \geq 3$ possess this strongest form of nonlocality.
2. New Theoretical Framework: The link between full nonlocality and antidistinguishability provides a powerful new tool for analyzing quantum correlations. It shifts the problem from constructing complex Bell inequalities to analyzing the geometric properties (overlaps) of post-measurement states.
3. Practical Applications:
   • Device-Independent Protocols: Fully nonlocal correlations are superior resources for Device-Independent Quantum Key Distribution (DI-QKD) and randomness certification because they are robust against noise and do not require high detection efficiencies to violate inequalities.
   • Quantum Advantage: The results suggest that shallow quantum circuits utilizing non-maximally entangled states could achieve computational advantages previously thought to require maximally entangled states.
4. Activation Phenomenon: The demonstration that any pure entangled state can be activated to full nonlocality with multiple copies suggests that full nonlocality is a ubiquitous resource in quantum theory, accessible through entanglement concentration or multi-copy protocols.

Conclusion

Renner et al. fundamentally expand the understanding of quantum nonlocality. By establishing a rigorous connection to antidistinguishability, they demonstrate that full nonlocality is a generic feature of high-dimensional entangled states (even non-maximally entangled ones) and can be activated in all pure entangled states, while simultaneously characterizing the specific boundaries where local content persists.