Complexity rank one implies real rank zero

Qingnan An; Zhichao Liu

REVIEW 2 major objections 4 minor 14 references

Complexity rank one implies real rank zero

T0 review · 2 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Every unital $C^*$-algebra with complexity rank at most one has real rank zero.

desk verdict Complexity rank one implies real rank zero is likely correct, but the printed proof leans entirely on an unproved and ill-quantified Proposition 1.7, so the paper needs revision, not rejection. read the letter →

arxiv 2608.12089 v1 pith:ORUO53DL submitted 2026-08-12 math.OA

Qingnan An , Zhichao Liu This is my paper · ORCID

classification math.OA MSC 46L0546L8046L85

keywords complexityrankrealzerodecomposabilityfinite-dimensionalapproximationuniformRoealgebraalmostcommutingself-adjointmatricesspectral-separationestimatethree-cornerestimation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proves that any unital $C^*$-algebra that decomposes over finite-dimensional algebras in the sense of complexity rank at most one has real rank zero: its self-adjoint elements with finite spectrum form a norm-dense set. This establishes a previously open implication and, as a corollary, gives the uniform Roe algebra $C^*_u|\mathbb{Z}|$ real rank zero. The proof is an elementary construction: for every self-adjoint contraction and every tolerance, it builds a finite-dimensional subalgebra whose self-adjoint part contains an element within tolerance. Because that approximating element lives in a finite-dimensional subalgebra, it automatically has finite spectrum, which is exactly what real rank zero requires.

What carries the argument

The load-bearing mechanism is the three-corner estimation of Lemmas 3.2 and 3.3. Given the rank-one decomposition data $(C,D,E,h)$, Lemma 3.2 uses the almost-commuting self-adjoint matrices theorem inside the finite-dimensional algebra $E$ to replace $h$ by a positive contraction $h_0$ that commutes with an element $e_0\in E$ close to $h(1-h)a$. Lemma 3.3 then applies the spectral projections $p_0 = \mathbf{1}_{[0,1/3)}(h_0)$, $p_\diamond = \mathbf{1}_{[1/3,2/3]}(h_0)$, and $p_1 = \mathbf{1}_{(2/3,1]}(h_0)$: the middle corner is handled because $t(1-t)\ge 2/9$ there, and the two off-diagonal corners are controlled by the spectral-separation estimate of Lemma 2.8, whose proof uses exponential estimates and vector-valued integration. The diagonal corners are then approximated in the three algebras $D$, $E$, and $C$, and orthogonality assembles them into one finite-dimensional subalgebra.

What would settle it

Construct a unital $C^*$-algebra with complexity rank at most one in the original approximate sense of Definition 1.4 whose self-adjoint elements with finite spectrum are not dense; that would directly refute Theorem 4.1. The most concrete place to look is the uniform Roe algebra $C^*_u|\mathbb{Z}^2|$, which is known not to have real rank zero; if it could be shown to admit a rank-one decomposition, the theorem would be false. Short of a counterexample, one could check whether Proposition 1.7 is valid for finite-dimensional classes by examining the cited proof in [8], since that proposition is the load-bearing step.

Watch

Extended reading notes

Core claim

Theorem 4.1 states that if $A$ is a unital $C^*$-algebra with complexity rank at most one, then $A$ has real rank zero. The proof fixes a self-adjoint contraction $a$ and $\varepsilon>0$, obtains from the refined rank-one property a triple of unital finite-dimensional subalgebras $C,D,E\subseteq A$ with $E\subseteq C\cap D$ and a positive contraction $h\in E$ that almost commutes with $a$ and separates it into $ha$, $(1-h)a$, and $h(1-h)a$, and then perturbs $h$ to a positive contraction $h_0$ that commutes with a self-adjoint $e_0\in E$ approximating $h(1-h)a$. Cutting by the spectral projections of $h_0$ at $1/3$ and $2/3$ produces three diagonal corners that are approximated inside $D$, $E$, and $C$ respectively, while the off-diagonal blocks are controlled by an inversion on the middle corner and a spectral-separation estimate on the outer corners. The assembled element $b$ lies in the finite-dimensional subalgebra $p_0 D p_0 \oplus p_\diamond E p_\diamond \oplus p_1 C p_1$ and satisfies $\|a-b\|<\varepsilon$.

Load-bearing premise

The proof leans on an unproved strengthened form of rank-one decomposability that forces the small overlap algebra $E$ to lie exactly inside both large algebras $C$ and $D$; the corner approximations are built inside $E$ and promoted to $C$ or $D$ using exactly that inclusion, so without it the construction breaks.

Editorial extensions

If this is right

The uniform Roe algebra $C^*_u|\mathbb{Z}|$ has real rank zero (Theorem 4.4).
Since $C^*_u|\mathbb{Z}^2|$ has complexity rank at most two but lacks real rank zero, its complexity rank is exactly two (Remark 4.5).
The result answers Question 3.10 of [9] and partially answers Question 6.5 of [8].
If the universal-coefficient-theorem hypothesis from [14] is met for a unital purely infinite simple nuclear $C^*$-algebra with zero $K$-theory, that algebra decomposes over finite-dimensional algebras and therefore has real rank zero.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The proof's reliance on the exact inclusion $E\subseteq C\cap D$ in Proposition 1.7 is the point most likely to be probed; if that strengthened form of rank-one decomposability failed in a concrete example, the theorem might still be true but would require a different argument.
Because the constants in Lemmas 3.2 and 3.3 are explicit, the construction could in principle yield quantitative finite-dimensional approximations for concrete rank-one algebras such as $C^*_u|\mathbb{Z}|$.
The three-corner cut at $1/3$ and $2/3$ suggests a template for higher complexity ranks, where an algebra of rank $n$ might be approximated by a block decomposition with more corners; the paper does not attempt this iteration.
The related question of whether weak complexity rank one implies real rank zero remains open precisely because the strengthened Proposition 1.7 is not available in that setting.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: the result is new and the main body of the proof is in good shape, but the paper currently does not prove the one proposition that carries the weight. Proposition 1.7 is stated as the strengthened rank-one form, with condition (i) ∥[h,x]∥ < η and no quantifier on η. Theorem 4.1 applies it with a fixed small δ and needs ∥[h,a]∥ < δ. As printed that step does not follow; this looks like a typo rather than a fatal flaw, but it must be fixed. The exact inclusion E ⊆ C∩D is also asserted as the “substantive reduction” and attributed to [8, Proposition 2.14] without proof in this paper. The corner constructions and estimates in Lemmas 3.2 and 3.3 use that inclusion essentially: p0Dp0, p1Cp1, and p⋄Ep⋄ all depend on the projections lying in E, which then lies in C and D. I have not checked [8] to see whether the precise statement with exact inclusions is actually proved there; the authors need to give a direct proof or a precise reference.

What is genuinely good: Lemma 2.8 is a clean Rosenblum-type Sylvester estimate, and the boundary constants in Lemmas 3.2 and 3.3 check out. The three-corner approximation idea is nice: cut by h0, control outer off-diagonal blocks with Lemma 2.8 and the middle row with invertibility of p⋄k0p⋄, then approximate each diagonal corner with the corresponding corner element. The proof is not circular and does not fit parameters to an output; the dependence on Lin’s theorem is external and standard. The application to C_u^*|ℤ| follows if Theorem 4.1 holds, and the remark about ℤ² is a useful sanity check.

The acknowledgements disclose AI assistance for exposition; that is a disclosure norm issue, not a mathematical one.

Who this is for: operator algebraists working on complexity rank, decomposability, and real rank zero. A serious referee should take it. My advice: send it to peer review, but require the authors to quantify η properly and to prove Proposition 1.7, or cite the exact location in [8] and spell out why the strengthened form follows. If that proposition is true as intended, the main theorem is likely correct.

Referee Report

2 major / 4 minor

Summary. The paper claims that every unital C*-algebra with complexity rank at most one has real rank zero. The proof proceeds by taking an arbitrary self-adjoint contraction a and a tolerance ε>0, producing finite-dimensional unital subalgebras C,D,E of A satisfying a strengthened form of decomposability, and then using a three-corner perturbation argument. Lemmas 2.8, 3.2, and 3.3 build an exactly commuting pair in E and then a finite-dimensional subalgebra F with b∈F_sa and ∥a-b∥<ε. As a corollary, the authors derive that the uniform Roe algebra C_u^*|Z| has real rank zero.

Significance. If correct, this answers an open question in [8] (Question 6.5) and provides an elementary-looking proof of a structural property that had been related to other open problems in the area. The internal arguments in Sections 2 and 3 are carefully written: Lemma 2.8 is a correct Rosenblum-type estimate, Lemma 3.2's numerical bookkeeping checks out, and Lemma 3.3's corner decomposition is clear and the norm estimates are correct. The principal weakness is that the paper's central theorem is conditional on Proposition 1.7, which is cited from [8, Proposition 2.14] but not proved here and, as printed, is not even a well-formed statement because of the unquantified parameter η. The main proof is sound only if Proposition 1.7 is repaired and proved in the strong form used in Theorem 4.1.

major comments (2)

[Section 1, Proposition 1.7; Section 4, Theorem 4.1] The parameter η in condition (i) of Proposition 1.7 is never quantified. As written, the proposition has no well-defined truth value: it asserts existence of C,D,E,h such that ∥[h,x]∥<η, but η is not tied to the quantifiers. In Theorem 4.1 the authors apply Proposition 1.7 with X={a} and ε=δ, and then use Lemma 3.2, whose first hypothesis requires ∥[h,a]∥<δ. This forces η≤δ, or more generally η=η(δ) with η(δ)<δ. If η were independent of δ and large, the proof would not go through. The statement must be corrected, for example by quantifying η universally before the existential quantifiers or by giving an explicit function η(ε) with η(ε)<ε, and the proof of Theorem 4.1 must then verify the corresponding choice of δ.
[Section 1, Proposition 1.7(iii); Section 3, Lemmas 3.2 and 3.3] The exact inclusion E⊆C∩D in Proposition 1.7(iii) is strictly stronger than the approximate containment in Definition 1.4(iii), which only requires every element of the unit ball of E to be ε-close to C and to D. The authors cite [8, Proposition 2.14] and state that the passage to the strengthened form is 'the substantive reduction', but no proof of this reduction is included. The exact inclusion is genuinely load-bearing: in Lemma 3.3 the elements s0, s1, r⋄ and the corner algebras p0Dp0, p1Cp1, p⋄Ep⋄ can be formed only because p0,p1,p⋄ belong to E and E is contained in C∩D. Without a proof of Proposition 1.7 in its strong form, the main theorem is not established. I request that the authors either prove Proposition 1.7 (or reproduce its proof from [8] with the precise hypotheses) or restructure the argument so that it does not need this strengthened statement.

minor comments (4)

[Section 2, Definition 2.4] In Definition 2.4, 'sequece' should be 'sequence'.
[Section 3, Lemma 3.2] In the final estimate of ∥esa∥, the strict inequalities give ∥esa∥<1/2 rather than ≤1/2; the displayed '≤1/2' is harmless but should be made consistent.
[Section 3, Lemma 3.3] The 3-by-3 block matrix after equation (11) uses asterisks for controlled entries; adding explicit labels such as 'controlled by (11)' and 'controlled by (12)' inside the diagram or in the surrounding text would improve readability.
[Section 4, Remark 4.5] The phrase 'must has complexity rank two' should be 'must have complexity rank two'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the main gap is an ill-quantified cited proposition, not a circular argument.

full rationale

The paper's central claim is not circular. Theorem 4.1 assumes complexity rank at most one and proves real rank zero by an explicit perturbation argument. It does not use real rank zero as an input, nor does it fit any parameter to the quantity being predicted. The approximation b is constructed from auxiliary subalgebras C, D, E supplied by Proposition 1.7 and from Lin's almost-commuting matrix theorem (Theorem 3.1); Lemmas 2.8, 3.2 and 3.3 contain the actual derivation. Proposition 1.7 is cited from [8, Proposition 2.14], which is external work by Jaime and Willett and is not by the present authors; the exact inclusion E subseteq C intersect D is stronger than Definition 1.4, and the paper flags it as the substantive reduction. That is a reliance on an external theorem, not a self-citation or a definitional equivalence. The unquantified eta in condition (i) of Proposition 1.7 is a genuine rigor gap that affects the applicability of the proposition in Theorem 4.1, and this should be repaired; however, a missing quantification is not a circular reduction. No equation in the paper reduces to its own input by construction, so the circularity score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new free parameters or invented entities. Its central claim rests on a strengthened characterization of complexity rank one (Proposition 1.7) that is cited from prior work and is not proved in the paper. The ambiguous quantization of eta in Proposition 1.7 is an additional structural risk.

assumptions (4)

domain assumption Proposition 1.7: complexity rank at most one is equivalent to the existence of unital finite-dimensional C,D,E with E⊆C∩D and a positive contraction h in E satisfying conditions (i)-(iii), with a controlled commutator.
This is the load-bearing premise of Theorem 4.1. It is cited from [8, Proposition 2.14] but not proved here. The exact inclusion E⊆C∩D is essential for Lemmas 3.2 and 3.3.
standard math Theorem 3.1 (Lin's theorem): almost commuting self-adjoint matrices are close to commuting self-adjoint matrices.
Used in Lemma 3.2 to obtain commuting h0 and e0. It is a published result of Friis-Rørdam and Lin.
standard math Spectral mapping theorem and basic C*-algebra facts about corners and unital subalgebras.
Used throughout the proof, including Lemma 2.1, Lemma 2.8, and the corner constructions.
standard math Bochner integral facts and the fundamental theorem of calculus for norm-continuous functions.
Used in the proof of Lemma 2.8 to justify differentiating the exponential product.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Complexity rank one implies real rank zero." pith.science (2026). https://pith.science/paper/ORUO53DL

@misc{pith2026260812089,
  author       = {Pith},
  title        = {Pith review of: Complexity rank one implies real rank zero},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ORUO53DL}},
  note         = {Machine review of arXiv:2608.12089}
}

abstract

We show that $C^*$-algebras with complexity rank one have real rank zero, as an application, the uniform Roe algebra $C_u^*|\mathbb{Z}|$ has real rank zero.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 12 canonical work pages

[8]

Jaime and R

A. Jaime and R. Willett,Complexity rank for C ∗-algebras, M¨ unster J. Math.16(2023), 51–94. 1, 2, 11, 12

work page 2023
[1]

L. G. Brown and G. K. Pedersen, C ∗-algebras of real rank zero, J. Funct. Anal.99 (1991), no. 1, 131–149. 1

work page 1991
[2]

Diestel and J

J. Diestel and J. J. Uhl, Jr.,Vector Measures, Mathematical Surveys, vol. 15, American Mathematical Society, Providence, RI, 1977, xiii+322 pp., ISBN 978-0-8218-1515-1. 3

work page 1977
[3]

Friis and M

P. Friis and M. Rørdam,Almost commuting self-adjoint matrices–a short proof of Huaxin Lin’s theorem, J. Reine Angew. Math.479(1996), 121–132. 6

work page 1996
[4]

Guentner, R

E. Guentner, R. Tessera and G. Yu, A notion of geometric complexity and its application to topological rigidity, Invent. Math., 189 (2012), no. 2, 315–357. 1, 12

work page 2012
[5]

Guentner, R

E. Guentner, R. Tessera and G. Yu, Discrete groups with finite decomposition com- plexity, Groups Geom. Dyn., 7 (2013), no. 2, 377–402. 1

work page 2013
[6]

Guentner, R

E. Guentner, R. Willett and G. Yu, Finite dynamical complexity and controlled operator K-theory, arXiv:1609.02093v3. 1

arXiv
[7]

Hyt¨ onen, J

T. Hyt¨ onen, J. van Neerven, M. Veraar, L. Weis. Analysis in Banach Spaces Volume I: Martingales and Littlewood-Paley Theory. Ergebnisse der Mathematik und ihrer Grenzgebiete. 3. Folge / A Series of Modern Surveys in Mathematics (Springer Cham, 2023). 3

work page 2023

Show all 14 references

[9]

Li and R

K. Li and R. Willett, Low-dimensional properties of uniform Roe algebras, J. Lond. Math. Soc. (2), 97 (2018), no. 1, 98–124. 1, 12, 13

2018
[10]

Lin,Almost commuting selfadjoint matrices and applications, in P

H. Lin,Almost commuting selfadjoint matrices and applications, in P. A. Fillmore and J. A. Mingo (eds.),Operator Algebras and Their Applications, Fields Inst. Commun., vol. 13, Amer. Math. Soc., Providence, RI, 1997, 193–233, MR 1424963. 6

1997
[11]

Rosenblum,On the operator equation BX−XA = Q, Duke Math

M. Rosenblum,On the operator equation BX−XA = Q, Duke Math. J.23(1956), no. 2, 263–269, doi:10.1215/S0012-7094-56-02324-9. 5

1956 doi
[12]

ˇSpakula and R

J. ˇSpakula and R. Willett, A metric approach to limit operators, Trans. Amer. Math. Soc., 369 (2017), no. 1, 263–308

2017
[13]

S. Wei. On the quasidiagonality of Roe algebras, Sci. China Math., 54 (2011), 1011– 1018

2011
[14]

Willett and G

R. Willett and G. Yu. The Universal Coefficient Theorem for C ∗-Algebras with Finite Complexity. Memoirs of the European Mathematical Society, Vol. 8, 2024 1, 2, 12 School of Mathematics and Statistics, Northeast Normal University, Changchun, 130024, China Email address:qingna...

2024

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [8]

Jaime and R

A. Jaime and R. Willett,Complexity rank for C ∗-algebras, M¨ unster J. Math.16(2023), 51–94. 1, 2, 11, 12

work page 2023

[2] [1]

L. G. Brown and G. K. Pedersen, C ∗-algebras of real rank zero, J. Funct. Anal.99 (1991), no. 1, 131–149. 1

work page 1991

[3] [2]

Diestel and J

J. Diestel and J. J. Uhl, Jr.,Vector Measures, Mathematical Surveys, vol. 15, American Mathematical Society, Providence, RI, 1977, xiii+322 pp., ISBN 978-0-8218-1515-1. 3

work page 1977

[4] [3]

Friis and M

P. Friis and M. Rørdam,Almost commuting self-adjoint matrices–a short proof of Huaxin Lin’s theorem, J. Reine Angew. Math.479(1996), 121–132. 6

work page 1996

[5] [4]

Guentner, R

E. Guentner, R. Tessera and G. Yu, A notion of geometric complexity and its application to topological rigidity, Invent. Math., 189 (2012), no. 2, 315–357. 1, 12

work page 2012

[6] [5]

Guentner, R

E. Guentner, R. Tessera and G. Yu, Discrete groups with finite decomposition com- plexity, Groups Geom. Dyn., 7 (2013), no. 2, 377–402. 1

work page 2013

[7] [6]

Guentner, R

E. Guentner, R. Willett and G. Yu, Finite dynamical complexity and controlled operator K-theory, arXiv:1609.02093v3. 1

arXiv

[8] [7]

Hyt¨ onen, J

T. Hyt¨ onen, J. van Neerven, M. Veraar, L. Weis. Analysis in Banach Spaces Volume I: Martingales and Littlewood-Paley Theory. Ergebnisse der Mathematik und ihrer Grenzgebiete. 3. Folge / A Series of Modern Surveys in Mathematics (Springer Cham, 2023). 3

work page 2023