{"id":"3e2e7f04-6f57-4397-a861-61f7bbe87006","arxiv_id":"2607.09580","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Strong constraints in the height-conserving quantum dimer model produce momentum-dependent level statistics while eigenstates still obey ETH-like entanglement in every resolved Krylov sector.","lead":"In a height-conserving quantum dimer model, after removing frozen states and resolving all conserved quantities, level statistics vary from near-Poisson to Wigner-Dyson by momentum sector, yet eigenstate entanglement always forms a thermal dome. This shows spectral chaos diagnostics and eigenstate thermalization can decouple under strong kinematic constraints.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"Finite-size and single-coupling limitations leave the claimed decoupling of spectral statistics from ETH untested in the thermodynamic limit.","rationale":"The Reader correctly isolates the single-system-size limitation as the weakest assumption underlying the strongest claim. The numerical evidence on 8\times8 is carefully symmetry-resolved and internally consistent, so the paper still merits a CONDITIONAL acceptance; the concern does not rise to a rejection. No stronger internal inconsistency (e.g., an algebraic error or a misidentified Krylov component) is apparent. The concrete test above is the minimal next calculation that would either solidify or refute the thermodynamic-limit interpretation of the reported decoupling.","tokens_in":14895,"tokens_out":535,"duration_ms":5703,"concrete_test":"Construct the dominant Krylov component for at least one of the three representative sectors on a 10\times10 torus (or a 8\times10 cylinder with the same height and winding quantum numbers), recompute the adjacent-gap ratio ⟨r⟩ and the energy-resolved entanglement entropy for the corresponding momentum blocks, and check whether (i) the spread of ⟨r⟩ across momenta remains of order 0.1 and (ii) the entanglement dome stays narrow and free of additional low-entanglement outliers. If either diagnostic collapses toward a single universal value, the claimed asymptotic decoupling is not supported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that, after resolving all six conserved quantities plus momentum and discarding frozen states, the dominant Krylov components of the height-conserving quantum dimer model exhibit sector-dependent level statistics (near-Poisson to GOE) while the eigenstate entanglement entropy remains uniformly dome-shaped and ETH-like. All numerical support for this decoupling is obtained on a single 8\times8 lattice at the single ratio v/t = −0.5. The paper itself states that “the limited system size prevents a systematic finite-size scaling.” Because the Hilbert-space dimension of each momentum block is only a few thousand, residual finite-size correlations, incomplete unfolding, or residual phase coherence induced by the height constraints can still produce intermediate ⟨r⟩ values that would drift toward either pure Poisson or pure GOE once the linear size is increased. Without at least one larger lattice (or a controlled extrapolation of both ⟨r⟩ and the width of the entanglement dome) the observed decoupling cannot be asserted to survive in the thermodynamic limit, which is precisely the regime in which “constrained quantum chaos” is claimed to be a robust phenomenon.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript studies thermalization in the height-conserving quantum dimer model (hQDM) on the square lattice. After resolving the two winding numbers, four sublattice height charges, and lattice momentum, and after discarding frozen configurations, the authors isolate the dominant connected Krylov component of each symmetry sector. On an 8\times8 lattice at v/t = −0.5 they compute adjacent-gap ratios, spectral form factors, bipartite entanglement entropy, and graph-Laplacian connectivity for three representative classes of momentum sectors. They report that level statistics range from near-Poisson (⟨r⟩ ≈ 0.4–0.45) to GOE-like (⟨r⟩ ≈ 0.53) depending on the momentum block, while the eigenstate entanglement entropy collapses onto a narrow, dome-shaped energy-resolved curve in every sector, interrupted only by a handful of low-entanglement outliers in selected blocks. The authors interpret this as a concrete instance of constrained quantum chaos in which spectral correlations and ETH need not share the same universal signatures.","tokens_in":15158,"tokens_out":1064,"duration_ms":9813,"significance":"If the reported decoupling survives in the thermodynamic limit, the work supplies a clean, fully symmetry-resolved example in which strong kinematic constraints (height conservation plus dimer packing) produce momentum-dependent intermediate spectral statistics while the bulk of the eigenstates remain ETH-compliant. That would sharpen the distinction between level-repulsion diagnostics and eigenstate thermalization under Hilbert-space fragmentation and would be of interest to the communities working on constrained dynamics, quantum scars, and lattice gauge theories. The technical execution—explicit construction of the six-charge sectors, isolation of the giant Krylov component via graph connectivity, and simultaneous use of gap ratios, SFF ramps, and entanglement domes—is careful and standard. The principal limitation is that all evidence is obtained at a single system size and a single coupling; the paper itself notes that finite-size scaling is unavailable.","major_comments":[{"comment":"Results and Discussion (and the explicit caveat in the Introduction): every quantitative claim rests on a single 8\times8 lattice at the single ratio v/t = −0.5. Momentum-block dimensions are only a few thousand states. Intermediate ⟨r⟩ values (0.4–0.45) and the narrowness of the entanglement dome can still be finite-size artifacts of residual phase coherence or incomplete unfolding. Without at least one larger lattice, a controlled extrapolation of both ⟨r⟩ and the width of the SvN dome, or a second generic value of v/t, the assertion that spectral statistics and ETH decouple under strong constraints cannot be regarded as established in the thermodynamic limit—the regime in which “constrained quantum chaos” is claimed to be robust.","section":null},{"comment":"Diagnostics and Results (Case-I–III): the paper reports that graph connectivity (mean degree, relative degree fluctuation, single zero mode of the Laplacian) is essentially identical across momentum subsectors that nevertheless display markedly different ⟨r⟩. This is used to argue that momentum projection modulates spectral correlations without altering connectivity. The argument would be substantially stronger if the phase-order parameter Φ defined in Eq. (10) were evaluated and tabulated for the same subsectors; without that datum the claimed mechanism remains qualitative.","section":null}],"minor_comments":[{"comment":"Abstract and throughout: “spectral spectral statistics” and “near-Poisoon” are typos that should be corrected.","section":null},{"comment":"Model section, Eqs. (2)–(3): the plaquette-pair geometry is described only in words; a small schematic of the allowed resonance moves would make the height-conservation rule clearer to non-specialists.","section":null},{"comment":"Results, Case-II: the text states “⟨z⟩ ≈ 0.42−0.45” when referring to gap ratios; this is a clear slip for ⟨r⟩ and should be fixed.","section":null},{"comment":"Figures 2–4: the Savitzky–Golay smoothing window and the precise energy window used for the linear SFF ramp fit are not stated; both should be given so that the slope values can be reproduced.","section":null},{"comment":"References: the recent literature on spectral statistics in other constrained dimer and gauge models could be expanded slightly to place the intermediate ⟨r⟩ values in a broader context.","section":null}],"recommendation":"major_revision","confidential_remarks":"The central observation is interesting and the numerics are carefully done, but the single-size, single-coupling limitation is load-bearing for a claim about a new regime of constrained chaos. I would be comfortable recommending acceptance after the authors either (i) add data for at least one larger system or a second generic v/t, or (ii) substantially soften the thermodynamic-limit language and reframe the work as a finite-size demonstration. Without one of those changes the manuscript is better suited to a more specialized venue."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The new result is concrete and useful: after resolving the six height/winding charges plus momentum and discarding frozen states, the dominant Krylov blocks of the height-conserving quantum dimer model show momentum-dependent gap ratios (near-Poisson to GOE) while the bipartite entanglement of almost every eigenstate sits on a clean thermal dome. A few low-entanglement outliers appear only in selected sectors. That decoupling is not in the earlier model paper or the dimer-ETH literature they cite, and it is cleanly documented for three representative sectors on the 8\times8 lattice.\n\nWhat they do well is the bookkeeping. They construct the connectivity graph, isolate the giant component, project to Bloch sectors, and then run the standard suite (adjacent-gap ratio, time-averaged SFF, von Neumann entropy, Laplacian spectrum). The graph diagnostics confirm that the blocks they study are singly connected with modest degree fluctuations, so residual fragmentation is not the explanation. The figures are readable and the parameter choice v/t = −0.5 is away from the RK point.\n\nThe soft spot is exactly the one the stress-test flags: everything is L = 8 and one coupling. Hilbert-space dimensions per momentum block are only a few thousand, so intermediate ⟨r⟩ values could still drift. The authors themselves note that finite-size scaling is impossible. That limits how strongly one can claim “constrained quantum chaos” as an asymptotic phenomenon, but it does not make the observed decoupling on this lattice an artifact. The diagnostics are external and correctly applied; there is no circular redefinition of quantities.\n\nThis is for people already working on Hilbert-space fragmentation, quantum scars, or constrained chaos. It is not a broad re-organization of thermalization theory, but it is a solid numerical data point that a serious referee should see. I would send it out; the finite-size limitation is the main revision request, not a desk-reject reason. Worth a look if you are writing on related models.","headline":"Clean ED demonstration that level statistics and ETH can decouple inside fully resolved Krylov sectors of the height-conserving QDM; finite-size caveat is real but does not erase the result.","tokens_in":15727,"tokens_out":506,"would_cite":true,"duration_ms":5011,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"In a height-conserving quantum dimer model, eigenstates thermalize even when level statistics swing from Poisson-like to Wigner-Dyson across momentum sectors.","keywords":["quantum dimer model","height conservation","Hilbert-space fragmentation","eigenstate thermalization","level statistics","constrained quantum chaos","Krylov subspace"],"falsifier":"A systematic finite-size scaling of the entanglement dome width and of the adjacent-gap ratio on lattices larger than 8 by 8 that either collapses the dome or forces all momentum sectors into a single universal level-spacing distribution.","tokens_in":15807,"feed_emoji":"⚛️","tokens_out":741,"duration_ms":8847,"temperature":0.7,"pith_summary":"Strongly constrained quantum systems often fragment their Hilbert space, so it is natural to ask whether the remaining connected pieces still thermalize. This paper studies the height-conserving quantum dimer model on the square lattice, whose local plaquette flips conserve winding numbers and four sublattice heights. After stripping out frozen states and resolving all conserved labels plus lattice momentum, the authors examine the dominant Krylov component of each sector. Across those momentum blocks the level-spacing statistics range from near-Poisson to Wigner-Dyson, yet the bipartite entanglement entropy of every eigenstate collapses onto a single narrow, dome-shaped curve that signals eigenstate thermalization. Only a few low-entanglement outliers appear in selected sectors. The result shows that spectral correlations and eigenstate thermalization need not march in lockstep once strong kinematic constraints are present, giving a concrete instance of constrained quantum chaos.","feed_headline":"Constrained dimers thermalize even when spectra look Poisson","feed_subtitle":"Level statistics swing with momentum, yet every eigenstate still forms a thermal entanglement dome.","key_machinery":"The dominant connected Krylov component obtained by resolving the six conserved height-related quantities together with lattice momentum; this component isolates the ergodically reachable subspace in which both spectral statistics and entanglement are computed.","core_discovery":"After fully resolving winding numbers, four sublattice heights and lattice momentum, and after discarding frozen states, the dominant connected Krylov component of every examined sector of the height-conserving quantum dimer model still produces a smooth, dome-shaped entanglement-versus-energy curve characteristic of the eigenstate thermalization hypothesis, even though the same sectors display level-spacing statistics that range from near-Poisson to Wigner-Dyson.","pith_inferences":["The same decoupling of level statistics from ETH may appear in other height- or charge-conserving lattice gauge models once their dominant Krylov components are isolated.","If larger-system methods confirm the entanglement dome, constrained quantum chaos would become a practical design principle for platforms that need thermalization without full random-matrix spectral rigidity.","The momentum dependence of the gap-ratio distribution suggests that residual phase coherence, rather than graph connectivity, is the main control knob for spectral correlations."],"forward_implications":["Spectral diagnostics alone can no longer be trusted as a complete proxy for thermalization once kinematic constraints are strong.","Eigenstate thermalization can remain intact inside a fragmented sector after frozen states are removed.","Momentum projection can modulate level statistics without destroying Hilbert-space connectivity or the thermal character of eigenstates.","Many-body scars appear only as rare, sector-selective outliers rather than as a generic feature of the constrained dynamics."],"fun_headline_variants":["Constrained dimers form ETH domes despite mixed level stats","Height-conserving dimers thermalize after resolving sectors","Dimers show entanglement ETH even when spectra look Poisson","Constrained Krylov sectors yield thermal entanglement curves","Quantum dimers: spectra vary, yet eigenstates stay thermal"],"cache_read_input_tokens":128,"weakest_assumption_plain":"That the largest lattice accessible by exact diagonalization (8 by 8) already captures the asymptotic thermalizing behavior of these Krylov components, so the observed dome-shaped entanglement will survive in the thermodynamic limit.","fun_headline_variants_meta":{"raw":{"variants":["Constrained dimers form ETH domes despite mixed level stats","Height-conserving dimers thermalize after resolving sectors","Dimers show entanglement ETH even when spectra look Poisson","Constrained Krylov sectors yield thermal entanglement curves","Quantum dimers: spectra vary, yet eigenstates stay thermal"]},"model":"grok-4.5","effort":"low","cost_usd":0.00416,"raw_usage":{"total_tokens":1276,"prompt_tokens":778,"num_sources_used":0,"completion_tokens":58,"cost_in_usd_ticks":41600000,"prompt_tokens_details":{"text_tokens":778,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":440,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":778,"tokens_out":58,"duration_ms":4510,"temperature":1.0,"reasoning_tokens":440,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-13T01:59:00.577085+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A systematic finite-size scaling of the entanglement dome width and of the adjacent-gap ratio on lattices larger than 8 by 8 that either collapses the dome or forces all momentum sectors into a single universal level-spacing distribution.","supporting_citations":[],"review_version":1}