---
title: Trilocal Inequality in Quantum Networks
url: https://www.emergentmind.com/topics/trilocal-inequality
type: topic
---

# Trilocal Inequality in Quantum Networks

Searching arXiv for the cited papers to ground the article in current literature.
A trilocal inequality is a Bell-type constraint derived for a network with **three independent sources**, where the hidden-variable distribution factorizes across those sources. In contrast with standard Bell locality, the source-independence assumption generally makes the classical correlation set non-convex, so trilocal constraints are typically **nonlinear** and depend strongly on network topology, measurement structure, and symmetry reductions. The term is therefore used for a family of related inequalities: for linear four-party chains, five-party multipartite-source networks, triangle networks with fixed local measurements, star-shaped four-party networks, asymmetric multisource configurations, and the fully symmetric no-input binary-output triangle network. Across these settings, violation of the relevant inequality certifies **nontrilocality**, i.e. network nonclassicality that cannot be explained by local response functions fed by three independent hidden variables [1411.4188][1707.00197][2303.08739][2503.16654].

## 1. Source independence and the meaning of trilocality

The common structural assumption is **factorization of hidden variables**. In the five-party scenario of Mukherjee et al., the trilocal model is
\[
P(a,b,c,d,e|x,y,z,w,u)
= \int d\lambda_1 d\lambda_2 d\lambda_3\,
\rho_1(\lambda_1)\rho_2(\lambda_2)\rho_3(\lambda_3)
\,P(a|x,\lambda_1)P(b|y,\lambda_1,\lambda_2,\lambda_3)
P(c|z,\lambda_1,\lambda_2,\lambda_3)P(d|w,\lambda_2)P(e|u,\lambda_3),
\]
with the strict independence condition
\[
\rho(\lambda_1,\lambda_2,\lambda_3)=\rho_1(\lambda_1)\rho_2(\lambda_2)\rho_3(\lambda_3)
\]
[1707.00197].

In the linear \(n\)-local framework specialized to \(n=3\), the four-party conditional probability likewise factorizes over three sources,
\[
P(a_1,a_2,a_3,a_4|x_1,x_2,x_3,x_4)
=\int d\lambda_1 d\lambda_2 d\lambda_3\,
\rho_1(\lambda_1)\rho_2(\lambda_2)\rho_3(\lambda_3)\,
P(a_1|x_1,\lambda_1)P(a_2|x_2,\lambda_1,\lambda_2)
P(a_3|x_3,\lambda_2,\lambda_3)P(a_4|x_4,\lambda_3)
\]
[1411.4188].

In triangle-network work, the same principle appears in a no-input form. For the fully symmetric binary-output triangle network, a distribution is triangle-local, or “trilocal,” iff
\[
p(a,b,c)=\sum_{\alpha,\beta,\gamma} q(\alpha)r(\beta)s(\gamma)\,
A(a|\beta,\gamma)\,B(b|\gamma,\alpha)\,C(c|\alpha,\beta),
\]
with independent sources \(q,r,s\) and deterministic response functions \(A,B,C\in\{0,1\}\) [2503.16654].

A central conceptual consequence is that the relevant classical sets differ from ordinary Bell-local polytopes. In standard Bell scenarios the local set is convex, whereas in network scenarios the trilocal set is generally non-convex, and the corresponding witnesses become nonlinear or algebraic rather than linear [2503.16654].

## 2. Canonical formulations across network topologies

The phrase “trilocal inequality” does not denote a single universal formula. It denotes the inequality appropriate to a specified three-source network.

| Scenario | Hidden-variable structure | Representative trilocal inequality |
|---|---|---|
| Linear four-party chain [1411.4188] | \(S_1,S_2,S_3\) in a line | \(\sqrt{|I|}+\sqrt{|J|}\le 1\) |
| Five-party multipartite-source network [1707.00197] | three independent tripartite sources | \(\sqrt[3]{|I_{i_1,i_2,0}|}+\sqrt[3]{|I_{j_1,j_2,1}|}\le 1\) |
| Triangle with fixed local measurements [2303.08739] | three pairwise sources, no inputs, four outcomes | \(\sqrt{|I_1|}+\sqrt{|I_2|}\le 1\) |
| Four-party star network [2603.20035] | central node and three edge links | \(\frac12\sum_{i=1}^4 |J_i|^{1/3}\le 2^{1/3}\) |
| Asymmetric four-party networks [2311.04583] | central Bob and three edge parties | \(T_3\le 2\cdot 6^{2/3}\), \(T'_3\le 2\cdot 15^{1/3}\) |
| Fully symmetric no-input triangle [2503.16654] | three pairwise sources, binary outputs | six boundary inequalities \(G,G',W_1,\dots,W_5\) plus bit-flips |

In the linear four-party chain, the correlators are
\[
E_{x_1x_2x_3x_4}
=\sum_{a_1\ldots a_4=0,1}(-1)^{a_1+a_2+a_3+a_4}P(a_1,a_2,a_3,a_4|x_1,x_2,x_3,x_4),
\]
and one defines
\[
I \equiv \frac14\,[E_{0,0,0,0}+E_{0,0,0,1}+E_{1,0,0,0}+E_{1,0,0,1}],
\]
\[
J \equiv \frac14\,[E_{0,1,1,0}-E_{0,1,1,1}-E_{1,1,1,0}+E_{1,1,1,1}],
\]
after which the trilocal bound is
\[
\sqrt{|I|}+\sqrt{|J|}\le 1
\]
[1411.4188].

In the five-party network of Mukherjee et al., one introduces eight grouped correlators
\[
I_{i,j,k}
= \frac18 \sum_{x,w,u=0}^1 (-1)^{k(x+w+u)}\,
\langle A_x B_i C_j D_w T_u\rangle,
\]
and for all \(i_1,i_2,j_1,j_2\in\{0,1\}\) obtains
\[
\sqrt[3]{\,|I_{i_1,i_2,0}|\,}+\sqrt[3]{\,|I_{j_1,j_2,1}|\,}\le 1.
\]
The summary states that this yields \(4\times 4=16\) independent inequalities [1707.00197].

In the triangle-network criteria of Ghosh et al., each party performs a fixed local measurement with four possible outcomes, encoded by bit pairs. The resulting theorem is
\[
\sqrt{|I_1|}+\sqrt{|I_2|}\le 1,
\]
where \(I_j\) are linear combinations of correlators built from observables \(C_X^0,C_X^1\) [2303.08739].

In the symmetric no-input triangle network, the distribution is reduced to the symmetrized correlators
\[
E_1\equiv \langle A\rangle,\qquad
E_2\equiv \langle AB\rangle,\qquad
E_3\equiv \langle ABC\rangle,
\]
with
\[
p(a,b,c)=\frac18\,[1 + (a+b+c)E_1 + (ab+bc+ca)E_2 + (abc)E_3],
\]
subject to the positivity tetrahedron. The boundary then breaks into GHZ, \(W\), and \(\overline W\) lobes, and the local region is tested by a family of algebraic inequalities rather than by a single root inequality [2503.16654].

## 3. Derivation methods

A common derivational pattern is to start from the trilocal factorization, convert the observable of interest into products of single-source or single-party expectations, and then apply absolute-value bounds together with Hölder-type or related inequalities.

For the linear \(n=3\) chain, one inserts the factorized model into \(I\) and \(J\), integrates out the middle source contributions using \(|K|\le 1\), and then uses a two-term Hölder/triangle argument to obtain \(\sqrt{|I|}+\sqrt{|J|}\le 1\). The same work explicitly contrasts this with the ordinary Bell-local bound
\[
|I|+|J|\le 1,
\]
which no longer uses source factorization [1411.4188].

For the five-party network, the derivation proceeds by introducing local expectations such as \(\langle A_x\rangle_{\lambda_1}\), bounding grouped correlators by successive applications of \(|mnp|\le |m||n||p|\), and then applying the inequality
\[
\sqrt[3]{m m' m''}+\sqrt[3]{s s' s''}\le
\sqrt[3]{(m+s)(m'+s')(m''+s'')}
\]
together with the dichotomic constraint \((|A_0\pm A_1|/2+|A_0\mp A_1|/2)\le 1\). This yields the cube-root trilocal inequality [1707.00197].

For the triangle-network inequalities with four outcomes and no inputs, the proof outline uses factorization at the \(\lambda\)-level and then the elementary inequality
\[
\sqrt{ab}+\sqrt{a'b'}\le \sqrt{(a+a')(b+b')},\qquad a,b,a',b'\ge 0,
\]
to derive
\[
\sqrt{|I_1|}+\sqrt{|I_2|}\le 1
\]
[2303.08739].

The derivation in asymmetric four-party trilocal scenarios is more elaborate. Under the factorized hidden-variable model, the correlators are bounded by repeated use of a product-sum inequality, and the optimal quantum values are obtained by a **sum-of-squares** construction. The paper constructs a positive-semidefinite operator
\[
\Gamma = \sum_j (\omega_j)^\alpha M_j^\dagger M_j
\]
with \(\alpha=1/3\) for \(T_3\) or \(\alpha=1/2\) for \(T'_3\), and chooses \(M_j\) so that \(\langle\Gamma\rangle=\beta-(T)_Q\ge 0\). The SOS method then gives the optimal quantum bounds without assuming the system dimension [2311.04583].

The symmetric binary-output triangle case follows a different route. Da Silva et al. numerically search for local models in carefully chosen two-dimensional affine planes, observe that optimal boundary models collapse to small two-parameter families with hidden-variable cardinalities \(\le 3\) in the symmetric case, and then eliminate the parameters using **resultant** or **Gröbner-basis** methods. The resulting master polynomial relations in \((E_1,E_2,E_3)\) define the conjectured analytic boundary surfaces [2503.16654].

## 4. Quantum violations, thresholds, and robustness

Quantum violations are known in several trilocal scenarios. In the linear four-party chain, an entanglement-swapping chain of three singlets with Bell-basis measurements at the two middle parties and observables
\[
A^0 = (\sigma_z+\sigma_x)/\sqrt2,\qquad
A^1 = (\sigma_z-\sigma_x)/\sqrt2
\]
at the ends yields
\[
I_Q=-\frac12,\qquad J_Q=-\frac12,
\]
hence
\[
\sqrt{|I_Q|}+\sqrt{|J_Q|}=\sqrt2\simeq 1.414>1.
\]
For Werner sources \(\chi(\alpha)=\alpha|\psi^-\rangle\langle\psi^-|+(1-\alpha)\mathbb{1}/4\), the summary gives \(\alpha_{\text{triloc}}=1/2\), whereas the usual CHSH threshold is \(\alpha_{\text{loc}}=1/\sqrt2\simeq 0.707\). Thus the interval \(0.5<\alpha<0.707\) is non-trilocal but Bell-local [1411.4188].

In the five-party multipartite-source scenario, the GGHZ family
\[
|\phi_{GGHZ}\rangle=\cos\alpha\,|000\rangle+\sin\alpha\,|111\rangle
\]
leads to the maximal violation
\[
S_{\text{trilocal}}=\sqrt[3]{2}\,\sin 2\alpha.
\]
Therefore any GGHZ state with \(\sin 2\alpha>2^{-1/3}\) is nontrilocal, and the perfect GHZ state gives \(S=2^{1/3}\approx 1.26>1\). The same summary states that genuine entanglement content is not an essential requirement: a biseparable state can also display nontrilocality, and for noisy GHZ sources the condition is \(\epsilon_1\epsilon_2\epsilon_3>1/2\), corresponding to a symmetric threshold \(\epsilon_c=2^{-1/3}\) [1707.00197].

In the four-outcome triangle-network criteria, explicit quantum violations are given for three Bell states with an entangled measurement basis, where
\[
\sqrt{|I_1|}+\sqrt{|I_2|}
= \sqrt{|2\alpha^2-\beta^2|}
+ \sqrt{|\alpha^6-\alpha^4\beta^2+\beta^6+\alpha^2(3-\beta^4)|},
\]
and this exceeds \(1\) for \(\alpha\gtrsim 0.892\). The same work reports that the product basis \(\{|00\rangle,|01\rangle,|10\rangle,|11\rangle\}\) yields \(I_1=1/4\), \(I_2=1/2\), hence \(\sqrt{I_1}+\sqrt{I_2}>1\). For detector inefficiency modeled by POVM elements \(p_4|b_i\rangle\langle b_i|+(1-p_4)\mathbb{I}/4\), violation holds when \(p_4\) exceeds approximately \(0.9\) for an optimal \(\alpha\) [2303.08739].

In the star-shaped four-party setting used for full-network-nonlocality-based QKD, if each source distributes a two-qubit state with diagonalized correlation tensor \(T_i=\mathrm{diag}(t_{i,1},t_{i,2},t_{i,3})\), then the relevant expression is
\[
B_{(3\text{-star})}\equiv
\sqrt{(t_{1,1}t_{2,1}t_{3,1})^{2/3}+(t_{1,2}t_{2,2}t_{3,2})^{2/3}},
\]
and any trilocal model must satisfy \(B_{(3\text{-star})}\le 2^{1/3}\). For three singlets, \(B_{(3\text{-star})}=\sqrt2\simeq 1.414>2^{1/3}\simeq 1.26\), and the summary states that no known measurement gives a larger value than \(\sqrt2\) in that symmetric three-singlet scenario [2603.20035].

For asymmetric trilocal scenarios, the optimal quantum bounds are
\[
(T_3)_Q^{opt}=4[2\sqrt3(1+\sqrt2)]^{1/3}\simeq 8.12,
\qquad
(T'_3)_Q^{opt}=6,
\]
to be compared with trilocal bounds \(2\cdot 6^{2/3}\simeq 6.60\) and \(2\cdot 15^{1/3}\simeq 4.93\), respectively. Under white-noise admixture, the critical visibility is reported as \(v_c\simeq 0.92\) for Scenario I and \(v_c\simeq 0.82\) for Scenario II [2311.04583].

## 5. The symmetric triangle network and analytic boundary inequalities

The most detailed recent use of the term concerns the **minimal triangle network** with no inputs and binary outcomes. In the fully permutation-symmetric sector, the observed distribution is encoded by the three correlators \(E_1,E_2,E_3\), and the triangle-local set splits into three boundary lobes centered on the archetypal GHZ, \(W\), and \(\overline W\) points [2503.16654].

The GHZ lobe is described by a two-parameter family of local models of cardinality \(3\times 3\times 3\), leading to an inequality of the form
\[
G(E_1,E_2,E_3)\ge 0,
\]
where, defining
\[
\Delta\equiv 3+5E_1+E_2-E_3,
\]
one has
\[
G\equiv
8(1+E_1)^3[11 + 37E_1 +28E_1^2 -4E_1^3 +8E_2 +13E_1E_2 +E_2^2
-11E_3 -18E_1E_3 -3E_2E_3 +2E_3^2]
-\Delta^4
\]
\[
+\sqrt{\Delta^2-8(1+E_1)^3}\,
[\Delta^3-4(1+E_1)^3(7+11E_1+E_2-3E_3)].
\]
Its bit-flipped copy is
\[
G(-E_1,E_2,-E_3)\ge 0.
\]
The summary notes that the square root requires \(\Delta^2\ge 8(1+E_1)^3\) to be real; if this fails, the GHZ-lobe test is not applied, and one resorts to the \(W\)-type tests [2503.16654].

The \(W\) lobe is covered by five distinct two-parameter local-model families. Their inequalities are:

\[
(W_1)\qquad -E_1E_2+E_3+(1-E_1)\sqrt{(1+E_2)^2-4E_1^2}\ge 0,
\]

\[
(W_2)\qquad
8E_1^5 -15E_1^4 -16E_1^3E_2 +22E_1^3 +16E_1^2E_2^2 -2E_1^2E_2 -20E_1^2
-6E_1E_2^2 +12E_1E_2 +10E_1 -E_2^2 -6E_2
\]
\[
+ (E_1^2-2E_1+2)E_3^2
+ (6E_1^3 -8E_1^2E_2 -12E_1^2 +10E_1E_2 +10E_1 -8E_2)E_3 -1
\ge 0,
\]
provided \(E_1\ge 1/3\),

\[
(W_3)\qquad
2(1+E_1)E_3^3
+27(1-3E_1+4E_1^2+E_2-2E_1E_2)(1-E_1+2E_1^3+2E_2-E_1E_2+E_2^2)
\]
\[
+54E_1E_3(2-5E_1+5E_1^2+3E_2-4E_1E_2+E_2^2)
-9E_3^2(1-2E_1-6E_1^2+E_2+2E_1E_2)\ge 0,
\]
subject to three auxiliary polynomial domain constraints,

\[
(W_4)\qquad \text{a degree-8 inequality valid in its own sub-region},
\]

\[
(W_5)\qquad E_3\ge f(E_1,E_2),
\]
where \(f\) is defined implicitly as the minimal \(E_3\) compatible with the symmetry equations and positivity, equivalently by solving a two-variable Lagrange-multiplier system [2503.16654].

For triangle-locality, the classical bound is the same structural rule throughout: the displayed expression is nonnegative, and equality defines the analytic boundary. The bit-flipped counterparts \(E_1\to -E_1\), \(E_3\to -E_3\) cover the \(\overline W\) lobe. The summary further states that all five \(W_i\) must be satisfied by any triangle-local point in the \(W\) lobe, and that violation of all five signals \(W\)-type nonlocality [2503.16654].

A significant open point is quantum realizability. The current status reported there is that **no explicit symmetric quantum realization is known** to violate these exact inequalities, although inflation-based outer approximations of the no-signaling-plus-source-independence region leave small wedges where quantum nonlocality may yet appear [2503.16654].

## 6. Relation to Bell nonlocality, generalizations, and applications

Trilocal inequalities differ from ordinary Bell inequalities in both structure and operational meaning. In the four-party linear chain, ordinary locality yields the linear constraint \(|I|+|J|\le 1\), whereas trilocality yields the nonlinear bound \(\sqrt{|I|}+\sqrt{|J|}\le 1\) [1411.4188]. In the five-party scenario, the summary explicitly states that without source independence one recovers only linear five-party Bell inequalities, which are far less noise-tolerant [1707.00197].

The triangle-network criteria sharpen this distinction further. Ghosh et al. emphasize that their inequalities can be violated even when each source state is individually CHSH-local; for Bell-diagonal states of the form specified in their Eq. (31), the individual pairs are CHSH-local yet the triangle-network criterion \(\sqrt{|I_1|}+\sqrt{|I_2|}\le 1\) can still be violated for suitable \(\alpha\) [2303.08739]. This isolates a form of network nonlocality not reducible to standard Bell-CHSH nonlocality.

The literature also distinguishes **full network nonlocality** from pairwise Bell nonlocality. In the star-network QKD setting, violation of the trilocal inequality cannot be obtained if any single source is classical, even if the other two are maximally nonlocal. The summary therefore presents the test as a genuine four-party nonlocality witness requiring all three sources [2603.20035]. This notion underlies the security analysis of the protocol \(\mathcal{N}_4\), where a subset of runs is used to test the trilocal inequality and the remaining runs determine the quantum bit error rate \(Q\). For three identical two-qubit states, the reported threshold is
\[
Q_0 = 1-\frac{\sqrt2(1+2^{1/6})^3}{16}\simeq 15.49\%,
\]
while for three non-identical states the strongest bound is \(Q_0\simeq 13.74\%\). A CHSH-based comparison protocol gives \(Q_0'\simeq 14.64\%\) [2603.20035].

Several generalizations are explicit. Mukherjee et al. state that their proof extends to an \(n\)-local star-chain with \(n\) sources and \(2n-1\) parties, replacing cube roots by \(n\)-th roots and yielding \(4^{n-1}\) inequalities [1707.00197]. Ghosh et al. generalize the fixed-measurement triangle-network construction to any \(n\)-sided polygon, introducing inequalities
\[
\sqrt{|I_{1,n}|}+\sqrt{|I_{2,n}|}\le 1
\]
for cycle networks [2303.08739]. Choudhary et al. develop asymmetric trilocal scenarios with unequal numbers of settings on the edge parties and central node, and use SOS methods to compute optimal quantum violations in a dimension-independent way [2311.04583].

Taken together, these results show that “trilocal inequality” has evolved into a broad technical category: nonlinear constraints enforcing three-source independence, tailored to specific network geometries, and used both as foundational diagnostics of network nonclassicality and as operational witnesses in tasks such as entanglement witnessing, noise-threshold analysis, and network-based quantum key distribution [1411.4188][1707.00197][2303.08739][2603.20035].

Source: https://www.emergentmind.com/topics/trilocal-inequality