Entanglement Entropy of Free Fermions on Random Fractal Lattices

Arkadiusz Kosior; Jia Wang

Entanglement Entropy of Free Fermions on Random Fractal Lattices

T0 review · 3 major / 3 minor · reviewed 2026-07-15 · grok-4.5

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

Pith's one-line read On random fractal lattices, free-fermion entanglement scales as a pure power of Hausdorff dimension, without the usual log enhancement, and quench growth collapses under Hausdorff size and spectral time.

desk verdict Abstract-only free-fermion fractal paper with a clean geometric-control claim that is interesting but currently unverifiable. read the letter →

arxiv 2607.05611 v2 pith:ZGCIKKDJ submitted 2026-07-06 quant-ph cond-mat.dis-nncond-mat.quant-gascond-mat.str-el

Arkadiusz Kosior , Jia Wang This is my paper · ORCID

classification quant-phcond-mat.dis-nncond-mat.quant-gascond-mat.str-el

keywords entanglemententropyfreefermionsrandomfractallatticesHausdorffdimensionspectralarealawquantumquenchgeometricdisorder

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 argues that free fermions living on random fractal lattices, with no onsite disorder, have ground-state bipartite entanglement that is controlled almost entirely by the lattice’s Hausdorff dimension. Subregions defined by graph distance obey a clean power-law area law; the logarithmic correction that appears for free fermions on ordinary Euclidean lattices is absent over a wide range of fillings and fractal parameters. After a global quench from a simple checkerboard product state, the growth of entanglement admits an asymptotic scaling collapse in which subsystem size is measured by the Hausdorff dimension while time is measured by the spectral dimension, producing only logarithmically slow growth over a long intermediate window. The result is that purely geometric randomness—generated by a stochastic growth algorithm plus a tunable density of missing links—is enough to reshape both static entanglement structure and the speed of quantum-information spreading in a free-fermion system.

What carries the argument

Random fractal lattices generated by a stochastic growth algorithm, with a tunable missing-link probability p that independently varies Hausdorff and spectral dimensions while leaving the Hamiltonian free of onsite disorder; entanglement is then measured for subregions defined by graph distance.

What would settle it

Compute the entanglement scaling on the same lattices for a sequence of increasing system sizes and check whether the pure Hausdorff power law persists without logarithmic corrections and whether the quench data continue to collapse when rescaled only by Hausdorff size and spectral time; any systematic residual log or failure of collapse falsifies the claim.

Watch

Extended reading notes

Core claim

Over a broad parameter range, bipartite entanglement entropy of free-fermion ground states on random fractal lattices exhibits robust power-law scaling governed primarily by the Hausdorff dimension, consistent with a generalized area law without the logarithmic enhancement of Euclidean free fermions; after a global quench, entanglement growth admits an asymptotic scaling collapse controlled by Hausdorff dimension (size) and spectral dimension (time), remaining logarithmically slow over an extended intermediate window.

Load-bearing premise

That subregions defined by graph distance, together with the stochastic growth algorithm and random missing-link probability, cleanly isolate Hausdorff and spectral dimensions as the sole geometric controls without residual finite-size, boundary or algorithm-specific artifacts that could mimic the reported power laws and scaling collapse.

Share X Bluesky LinkedIn Reddit HN

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

The one thing to know is that this is an abstract-only computational paper claiming that geometric disorder alone—via random fractal lattices with tunable Hausdorff and spectral dimensions, no onsite disorder—gives free-fermion ground states a clean Hausdorff-governed power-law entanglement (generalized area law, no log enhancement) and a quench whose scaling collapse is controlled by Hausdorff size and spectral-dimension time, remaining logarithmically slow. That combination is the actual novelty; entanglement on fractals and free-fermion area laws are not new, but this specific isolation of pure geometry is.

What they appear to do well is set up a clean control axis: stochastic growth plus missing-link probability p to dial d_H and d_s independently of onsite randomness, graph-distance bipartitions, and free-fermion numerics that should be reproducible in principle. The claims are coherent and, if the full data hold, useful for people who care about entanglement structure and information speed in disordered or non-integer-dimensional systems. No circularity is obvious from the abstract; dimensions are geometric observables and entanglement is computed separately.

The soft spot is exactly the one the stress-test flags, and it is load-bearing: we cannot check whether graph-distance cuts, the growth algorithm, and p really isolate those two dimensions without residual finite-size, boundary, or algorithm artifacts that could fake the power laws and the collapse. No figures, system sizes, fitting procedures, error bars, or controls are available. Soundness is therefore provisional; confidence has to stay low until the full text and methods appear. That is not a manufactured flaw—it is simply the state of the evidence.

This is for people working on free-fermion entanglement, fractal or disordered lattices, and quench dynamics. It deserves a serious referee once the full paper is out; the idea is sharp enough that a desk reject would be premature. I would not cite it yet and would not bring the abstract alone to reading group, but I would accept it for peer review and look carefully at the finite-size analysis and the operational definitions of d_H and d_s.

Referee Report

3 major / 3 minor

Summary. The manuscript studies bipartite entanglement entropy of free (noninteracting) fermions on random fractal lattices generated by a stochastic growth algorithm, with Hausdorff and spectral dimensions tuned via a growth parameter and a missing-link probability p, without onsite disorder. For ground states at various fillings, subregions defined by graph distance are reported to exhibit robust power-law entanglement scaling governed primarily by the Hausdorff dimension, consistent with a generalized area law and without the logarithmic enhancement known for Euclidean free fermions. After a global quench from an uncorrelated checkerboard state, entanglement growth is claimed to admit an asymptotic scaling collapse controlled by Hausdorff dimension (subsystem size) and spectral dimension (time), remaining logarithmically slow over an extended intermediate window. The central message is that geometric randomness alone can produce nontrivial ground-state entanglement structure and slow information spreading in free-fermion systems.

Significance. If the reported isolation of Hausdorff and spectral dimensions as the sole geometric controls is substantiated, the work would be a useful contribution to entanglement scaling beyond integer-dimensional lattices: it would supply a clean free-fermion setting in which a generalized area law without log enhancement, and a two-dimensional (d_H, d_s) dynamical collapse, can be tested against Euclidean benchmarks. The combination of ground-state scaling and post-quench dynamics on the same family of random fractals is of interest for quantum-information spreading in disordered geometries. Strengths claimed in the abstract—parameter-tuned geometry without onsite disorder, power-law rather than log-enhanced scaling, and an asymptotic collapse—are in principle falsifiable and would merit attention if backed by finite-size data, error analysis, and controls.

major comments (3)

The central claim that bipartite entanglement is governed primarily by the Hausdorff dimension (generalized area law, no log enhancement) rests on subregions defined by graph distance and on lattices generated by a stochastic growth algorithm plus missing-link probability p. From the abstract alone it is not possible to verify that residual finite-size, boundary, or algorithm-specific artifacts have been ruled out; a load-bearing requirement is a documented finite-size scaling analysis (range of system sizes, extraction of d_H, comparison to Euclidean free-fermion log enhancement, and controls that vary growth parameter and p independently while holding other geometry fixed).
The claimed asymptotic scaling collapse after a global quench—subsystem-size dependence set by Hausdorff dimension and temporal evolution by spectral dimension, with logarithmically slow intermediate growth—is load-bearing for the dynamical part of the paper. Without access to the operational definitions of d_H and d_s, the collapse protocol, the intermediate-time window, or raw scaling plots, it is not possible to assess whether the collapse is unique to those two dimensions or could be mimicked by other effective exponents or by the checkerboard initial state and graph-distance bipartition.
The abstract asserts robustness 'over a broad parameter range' and consistency with a generalized area law. That claim requires explicit fitting procedures, reported exponents versus independently measured d_H, and a quantitative statement that logarithmic corrections are absent (or bounded) rather than merely subdominant within the accessible sizes. Until those elements are inspectable, the isolation of geometric dimensions as sole controls remains an uncheckable assumption rather than a demonstrated result.

minor comments (3)

Only the abstract is available for this review; section numbering, equations, tables, figures, and methods cannot be cited. A full manuscript with methods, finite-size data, and scaling plots is required for a definitive report.
When the full text is supplied, the stochastic growth algorithm, the precise definition of graph-distance subregions, the extraction of Hausdorff and spectral dimensions, and the quench protocol should be stated with enough detail for independent reproduction.
Notation for filling fractions, the missing-link probability p, and the growth parameter should be introduced consistently and tied to the reported dimension ranges.

Circularity Check

0 steps flagged · score 0.0 of 10

Abstract-only review: no circular construction detectable; Hausdorff/spectral dimensions are geometric inputs and entanglement is an independent free-fermion observable.

full rationale

Only the abstract is available, so no equations, definitions of the growth algorithm, operational extraction of d_H and d_s, finite-size scaling plots, or self-citations can be inspected. From the abstract alone the claimed chain is non-circular by construction: lattices are generated by a stochastic growth algorithm plus missing-link probability p; Hausdorff and spectral dimensions are measured geometric properties of those lattices; bipartite entanglement entropy of free-fermion ground states (and post-quench dynamics) is then computed independently for subregions defined by graph distance. Reporting that the resulting EE scales as a power of subsystem size controlled by d_H, and that quench dynamics admit a collapse controlled by d_H (size) and d_s (time), does not redefine either dimension in terms of the entanglement data, nor does it fit a parameter to EE and then re-label the fit as a prediction. No uniqueness theorem, ansatz smuggled via self-citation, or renaming of a known Euclidean free-fermion result is present in the abstract. Residual risk that algorithm parameters are tuned until dimensions match desired values is ordinary model-building, not circularity of the kind enumerated in the instructions. Score 0 is therefore the honest finding for an abstract-only review; any stronger circularity claim would require quoting equations or self-citations that are unavailable.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

Abstract-only: free parameters of the growth algorithm and missing-link probability p are the main tunable knobs; standard free-fermion and fractal-geometry assumptions are taken as given. No new particles or forces are introduced. Full numerical values and any fitted exponents are unavailable.

free parameters (2)

growth parameter (stochastic growth algorithm)
Tunes Hausdorff and spectral dimensions of the random fractal lattice; exact range and fitting procedure not given in abstract.
missing-link probability p
Additional control that adds random edges; used to vary dimensions while keeping the system free of onsite disorder.

assumptions (3)

domain assumption Noninteracting fermions on a tight-binding graph Hamiltonian defined by the random fractal lattice adjacency.
Standard free-fermion setup; entanglement computed from the correlation matrix of the filled single-particle states.
domain assumption Hausdorff and spectral dimensions extracted from the generated lattices fully characterize the geometric disorder relevant to entanglement scaling.
Abstract asserts that scaling is governed primarily by these two dimensions; assumes other geometric details are secondary.
ad hoc to paper Subregions defined by graph distance are the appropriate bipartitions for testing a generalized area law.
Choice of cut is natural on graphs but is a modeling decision that could affect observed scaling.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Entanglement Entropy of Free Fermions on Random Fractal Lattices." pith.science (2026). https://pith.science/paper/ZGCIKKDJ

@misc{pith2026260705611,
  author       = {Pith},
  title        = {Pith review of: Entanglement Entropy of Free Fermions on Random Fractal Lattices},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZGCIKKDJ}},
  note         = {Machine review of arXiv:2607.05611}
}

abstract

Random fractal lattices provide a geometrically disordered setting in which quantum correlations can be shaped by noninteger dimensionality rather than onsite randomness. We investigate the entanglement properties of noninteracting fermions on random fractal lattices generated by a stochastic growth algorithm. By varying the growth parameter and adding missing links with probability $p$, we tune the Hausdorff and spectral dimensions while keeping the system free of onsite disorder. For ground states at different fillings, we compute the bipartite entanglement entropy of subregions defined by graph distance and analyze its scaling with subsystem size. Over a broad parameter range, we find robust power-law behavior governed primarily by the Hausdorff dimension, consistent with a generalized area law and without the logarithmic enhancement familiar from Euclidean free fermions. We also study entanglement growth following a global quench from an uncorrelated checkerboard state and uncover an asymptotic scaling collapse in which the subsystem-size dependence is governed by the Hausdorff dimension, while the temporal evolution is governed by the spectral dimension. The resulting dynamics are logarithmically slow over an extended intermediate-time window. These results show that geometric randomness alone can generate both nontrivial ground-state entanglement structure and slow quantum-information spreading in free-fermion systems.