Relative Balancedness Condition

• Relative balancedness condition is a cross-disciplinary label for reference-dependent constraints that compare an object to a benchmark, such as a signless operator or partition.
• It is applied in diverse areas including spectral theory of simplicial complexes, social choice, deep linear networks, convex geometry, and even quantum state analysis.
• Its significance lies in providing a unified framework to assess symmetry, cancellation, and invariant counting in systems with varying formal structures.

“Relative balancedness condition” is not a standardized formal term across the cited literature. In most of the relevant papers, the phrase does not appear literally; instead, each field introduces a domain-specific balancedness predicate that compares an object to a reference structure, such as a signless operator, a candidate bipartition, a distinguished point $r$, a fixed end-to-end map $X$, equal-length factors, or a larger support $M\supseteq E$ (Fan et al., 2024, Wang et al., 2020, Bludov, 9 Dec 2025, Lindsey et al., 3 Nov 2025, Espinoza et al., 3 Feb 2026). This suggests that the expression functions best as a cross-disciplinary editorial label for conditions of symmetry, cancellation, equalization, or support-invariant counting, rather than as a single axiom with a uniform definition.

1. Terminological status and domain map

Across the literature, the nearest formal object to a “relative balancedness condition” depends on the ambient theory. The common pattern is balance measured against something external: a complement, a total sum, a partition, a fiber, a frequency, a conormal fiber, or an operator with signs removed.

Domain	Closest formal notion	Representative condition
Simplicial complexes	Balancedness of the incidence signed graph $B_i(K)$ (Fan et al., 2024)	$$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$ balanced, for $(i+1)$-path connected $K$
Matrix-weighted signed networks	Unique non-trivial balancing set (Wang et al., 2020)	$$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$
House allocation	Balancedness of rank frequencies (Long et al., 2021)	$$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$
Random partition models	Relative log-concavity / $B$-sequence comparison (Lee et al., 2022)	$X$0 concave, equivalently $X$1
Balanced subsets	$X$2-balancedness (Bludov, 9 Dec 2025)	$X$3
Deep linear networks	Balanced manifold / zero moment map (Lindsey et al., 3 Nov 2025)	$X$4
Cellular automata	Balancedness of finite-pattern preimage counts (Capobianco et al., 2015)	every $X$5 has $X$6 pre-images on $X$7
Multi-objective optimization	Interval balancing of vector sums (Glaßer et al., 2010)	$X$8

A notable negative fact is itself stable across several papers: the phrase “relative balancedness condition” is often absent, while a nearby formal notion is explicit. That is stated for simplicial complexes, social choice, matrix-weighted networks, house allocation, balanced subsets, $X$9-core games, and deep linear networks. The resulting usage is therefore contextual rather than canonical.

2. Spectral and geometric formulations

In simplicial-complex spectral theory, the nearest formal content is the balancedness of the $M\supseteq E$0-th incidence signed graph $M\supseteq E$1. For a simplicial complex $M\supseteq E$2, the $M\supseteq E$3-th up Laplacian and signless up Laplacian are

$M\supseteq E$4

The core theorem states

$M\supseteq E$5

and, if $M\supseteq E$6 is $M\supseteq E$7-path connected,

$M\supseteq E$8

Balancedness here means that every cycle in the signed graph has positive sign-product, equivalently that the signed graph is switchable to the all-positive one. The paper also gives the orientation-theoretic formulation: there exists a choice of orientations on $M\supseteq E$9- and $B_i(K)$0-faces such that every incidence sign becomes $B_i(K)$1 (Fan et al., 2024).

In convex-geometric topology, the exact analogue is explicit and literal: a subset $B_i(K)$2 is $B_i(K)$3-balanced if

$B_i(K)$4

The coefficient formulation uses the weight polytope

$B_i(K)$5

and $B_i(K)$6 is $B_i(K)$7-balanced iff there exists $B_i(K)$8 with $B_i(K)$9. Minimal balanced subsets are affinely independent and have a unique weight vector. When $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$0, the poset of balanced subsets excluding $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$1 has order complex homotopy equivalent to a sphere: $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$2 where $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$3 (Bludov, 9 Dec 2025).

In singularity theory and conormal geometry, the comparable construction is not called balancedness, but the same relative structure appears in the description of the relative conormal fiber when $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$4. The paper proves

$$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$5

where $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$6 is the image of the exceptional divisor of the blowup of the conormal space along the graph of $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$7, and $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$8 is the projective join. The join term is the new component absent in the $$\lambda_{\max}(\mathcal L_i^{up}(K))=\lambda_{\max}(\mathcal Q_i^{up}(K))\iff B_i(K)$$9 case. Combined with the dimension condition $(i+1)$0 and the infinitesimal Whitney $(i+1)$1 fiber condition, this yields Thom’s $(i+1)$2 condition (Gaffney et al., 2018).

3. Social choice, allocation, and cooperative games

In social choice theory, the relevant condition is balancedness of a social choice correspondence. If $(i+1)$3 is obtained from a profile $(i+1)$4 by a transposition pair $(i+1)$5 via individuals $(i+1)$6, meaning that $(i+1)$7 and $(i+1)$8 are adjacent in opposite orders for $(i+1)$9 and $K$0 and then swapped in both rankings, balancedness requires

$K$1

The condition is an invariance property under offsetting adjacent swaps. The paper emphasizes its interaction with tops-only and shows that for $K$2 and $K$3, a correspondence satisfying tops-only and balancedness is constant on all non-unanimous profiles. Within scoring rules, balancedness characterizes the Borda correspondence except in the exceptional case $K$4, where weights $K$5 still yield a balanced non-Borda correspondence (Kelly et al., 2018).

In house allocation, the term is again simply balancedness, now as an ex ante rank-frequency symmetry across agents. A mechanism $K$6 is balanced if for every pair of agents $K$7 and every rank $K$8,

$K$9

The paper explicitly interprets this as equal treatment at the design stage when all preference profiles are equally likely, and notes that the notion corresponds to order symmetry with respect to the uniform distribution on the preference space. Its main theorem states that a mechanism is balanced, efficient, and group strategy-proof if and only if it is a TTC mechanism from individual endowments or a TC mechanism with three brokers; if $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$0, only TTC from individual endowments remains (Long et al., 2021).

In infinite cooperative game theory, the closest analogue is $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$1-balancedness, which is balancedness relative both to the feasible coalition structure $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$2 and to the $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$3-core. With

$$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$4

$$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$5

a game is $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$6-balanced if

$$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$7

The generalized Bondareva-Shapley theorem then asserts

$$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$8

The paper distinguishes this from Schmeidler $$\bigcap_{e\in\mathcal E^{nb}}\operatorname{null}(\mathcal W(e))\neq\{\mathbf 0\}$$9-balancedness and shows that the latter is insufficient for $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$0-core nonemptiness (Bartl et al., 2022).

4. Network dynamics and deep linear models

In signed matrix-weighted consensus, classical structural balance is no longer the correct criterion for bipartite consensus. The replacement notion is the non-trivial balancing set (NBS), defined relative to a node bipartition $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$1. Its balancing set $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$2 consists of the edges whose signs must be negated to make the graph structurally balanced with respect to that partition. It becomes a non-trivial balancing set when

$$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$3

The foundational theorem states that existence of an NBS with division $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$4 and a nonzero subspace $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$5 satisfying

$$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$6

is equivalent to

$$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$7

For general connected matrix-weighted networks, bipartite consensus implies existence of a unique NBS. If the graph has a positive-negative spanning tree, then bipartite consensus holds if and only if the graph has a unique NBS; if no NBS is present, the system admits trivial consensus (Wang et al., 2020).

In deep linear networks, balancedness is defined by vanishing adjacent-layer Gram-matrix differences: $$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$8 Equivalently, with

$$\left|\{R:f(R)_i=R_i^k\}\right|=\left|\{R:f(R)_j=R_j^k\}\right|$$9

balancedness is the zero level set of the moment map,

$B$0

For a fixed end-to-end map

$B$1

the fiber is

$B$2

and the main variational theorem is

$B$3

where $B$4 is the balanced manifold. Thus balancedness is the exact condition selecting the minimum-$B$5 representatives within a fixed equivalence class of factorizations of the same $B$6. The regularizing flow on the fiber obeys

$B$7

so imbalance decays exponentially to zero (Lindsey et al., 3 Nov 2025).

5. Partitions, words, and symbolic systems

In Bayesian random partition models, the nearest formal “relative” notion is comparative balancedness between models, expressed by the reverse dominance order on cluster-size vectors and the relative log-concavity order on Gibbs weight sequences. For $B$8, the order

$B$9

means that $X$00 is more balanced than $X$01. A Gibbs model is balance-averse if

$X$02

and balance-seeking if the inequality is reversed. To compare two Gibbs models with weight sequences $X$03, the paper uses relative log-concavity: $X$04 equivalently

$X$05

with

$X$06

A central conclusion is that product-form exchangeability plus projectivity make infinitely exchangeable Gibbs partitions balance-averse, which the paper identifies with rich-get-richer behavior (Lee et al., 2022).

In symbolic dynamics, balancedness is formulated by comparing occurrence counts in equal-length factors. For a bi-infinite word $X$07 and a factor $X$08, balancedness on $X$09 means existence of $X$10 such that

$X$11

The paper proves that this is equivalent to finiteness of the discrepancy $X$12, and, in a minimal uniquely ergodic subshift, equivalent to the coboundary identity

$X$13

For minimal dendric subshifts, balancedness on letters is equivalent to balancedness on factors (Berthé et al., 2018).

For Arnoux-Rauzy and Brun words, the paper uses discrepancy relative to the frequency vector. If $X$14 is the frequency vector, then

$X$15

and finite balancedness is equivalent, up to constants, to bounded discrepancy: $X$16 The main sufficient condition is bounded strong partial quotients. For Arnoux-Rauzy words this means

$X$17

and for Brun words on $X$18 letters

$X$19

Under these conditions the word is $X$20-balanced (Delecroix et al., 2013).

In the more recent factor-balancedness framework, a factor $X$21 is $X$22-balanced in $X$23 if

$X$24

The word is factor-balanced if every $X$25 has some $X$26, and uniformly factor-balanced if a single $X$27 works for all $X$28. The paper shows that this is equivalent, up to constants, to bounded discrepancy

$X$29

and gives general $X$30-adic sufficient conditions applicable in particular to linearly recurrent words. It also characterizes Sturmian words and ternary Arnoux-Rauzy words that are uniformly factor-balanced as precisely those with bounded weak partial quotients (Espinoza et al., 3 Feb 2026).

6. Quantum-state and entanglement formulations

In multipartite quantum information, the paper generalizes balancedness from qubits to qudits of local dimension

$X$31

For qubits, balancedness means that on each site the local basis states $X$32 and $X$33 occur equally often in the computational-basis expansion. For qudits, the generalized condition is that on each site all local basis states

$X$34

occur with equal weighted multiplicities.

If $X$35 records the local basis value at site $X$36 in column $X$37, then for prime $X$38 the balancing condition is expressed by

$X$39

For general $X$40, the paper replaces the roots-of-unity formulation by pairwise occurrence matrices $X$41 and requires

$X$42

The multiplicity vector $X$43 is taken in positive integers and may be interpreted as repeating the $X$44-th basis-product term $X$45 times.

The paper distinguishes balanced, partly balanced, and irreducibly balanced $X$46-orbits. Every stochastic state, meaning a pure state with local reduced density matrices $X$47, is balanced. Product states are not irreducibly balanced. Every irreducibly balanced state of length $X$48 is $X$49-equivalent to a stochastic state, and every irreducibly balanced state is stable, hence not in the SLOCC zero-class (Osterloh, 2014).

7. Algorithmic and operational formulations

In tree-to-word transducers over a bracket alphabet $X$50, balancedness is analyzed via reduction in the Dyck language. A normalized $X$51-copy tree-to-word transducer with axiom $X$52 is balanced precisely when its output language is $X$53. The decisive criterion is: $X$54 Here well-formedness means that every reduced output belongs to $X$55, and equivalence is tested after inversion and reduction. The paper also proves that well-formedness of a context-free language is decidable in polynomial time (Löbel et al., 2019).

In cellular automata over groups, balancedness is an exact finite-pattern preimage law. For a finite set $X$56, a larger finite set $X$57, and a pattern $X$58, balancedness requires exactly

$X$59

pre-images of $X$60 on $X$61. In the finitely generated case, if the neighborhood lies in a disk $X$62, this is equivalent to saying that every pattern on $X$63 has exactly

$X$64

pre-images on $X$65. Balancedness is equivalent to preservation of the uniform product measure on $X$66, and every reversible cellular automaton is balanced (Capobianco et al., 2015).

In multi-objective optimization, the closest formal object is an interval-balancing lemma for bounded vector sequences. For $X$67 with $X$68, there exist $X$69 intervals whose union $X$70 satisfies

$X$71

componentwise. Equivalently, $X$72 is close to one half of the total sum. The mixed two-sequence version yields

$X$73

again componentwise. The paper interprets this as a polynomial-time computable balance of two alternatives with conflicting costs and applies it to obtain a randomized $X$74-approximation for multi-objective maximum asymmetric traveling salesman and a deterministic $X$75-approximation for multi-objective maximum weighted satisfiability (Glaßer et al., 2010).

Taken together, these formulations show that “relative balancedness condition” is best understood as a family of reference-dependent constraints. In some settings the reference is spectral equality, in others a candidate partition, a half-total benchmark, a frequency vector, a fixed parameterization fiber, or a finite support enlargement. The phrase is therefore encyclopedically useful only when attached to its host theory; stripped of that context, it has no single invariant meaning.