{"id":"cb0aebbe-f75a-4659-ba48-7ee48c3a98f4","arxiv_id":"2411.14965","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Periodic graphs with infinitely many neighbours per vertex can have purely singular continuous spectrum, flat bands with only delocalized eigenvectors, and ballistic transport without dispersive estimates.","lead":"This paper studies infinite periodic graphs in which a single vertex can connect to infinitely many others, as long as the edge weights shrink fast enough to add up. It shows such graphs can have spectra and wave dynamics that are impossible for ordinary crystals, including waves that travel ballistically but never flatten out.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the central counterexample is internally valid; the Euclidean-coordinate definition of transport is an explicit modeling choice, not a gap.","rationale":"The reader's weakest assumption correctly identified the Euclidean-coordinate definition as the main caveat. I concur that it is a modelling choice rather than a flaw: the paper explicitly defines transport via the position operator x in §7.1.2 and justifies it in Remark 7.4. The exact counterexample to dispersion does not involve this choice, as it uses the L¹→L∞ norm. All key theorems are internally consistent; the proofs are explicit and parameter-free. The unproven genericity of the jump-set condition in Theorem 7.2 is an acknowledged assumption and does not affect the ν=1 examples of Theorem 1.1. Hence no objection that would change the verdict.","tokens_in":45715,"tokens_out":26120,"duration_ms":239901,"concrete_test":"As a verification step, compute the graph distance d(0,n) for the graph of Theorem 1.1(2) and confirm d(0,n)≤2 for all n∈Z; this validates Remark 7.4 that graph-distance transport would be degenerate. Independently, re-evaluate the exact integral (7.8) for a sample of times t outside 2πN to confirm the lower bound ‖e^{−itH}δ_0‖_∞ ≥ 1/π.","verdict_should_be":"UNCHANGED","load_bearing_attack":"We find no load-bearing flaw in the central claim. Theorem 1.1(2) is supported by exact computations: the Floquet function b(θ)=|1/2−θ| is in the Wiener algebra with nonnegative symmetric summable coefficients, b′≠0 a.e. gives pure absolutely continuous spectrum, the kernel formula (7.8)–(7.9) yields ‖e^{−itH}δ_0‖_∞ ≥ 1/π for all t, and Corollary 7.7 gives a nonzero ballistic speed for any state with finite second moment. The only caveat is the choice of the Euclidean coordinate operator x for transport, which is explicitly justified in Remark 7.4: for non-locally finite graphs such as Theorem 1.1(1) the graph distance is degenerate (d(0,n)=1 for all n), and even in the odd-supported graph (2) one has d(0,n)≤2, so graph-distance transport would not measure spreading. This is a modelling decision, not an inconsistency; the paper scopes its negative answer to Open Question 9 to this setting.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Z^d-periodic weighted graphs with finite fundamental cell and possibly infinite vertex degree, under nonnegative, symmetric, summable edge weights. It develops the Floquet theory of such non-locally finite crystals and exhibits a collection of exotic spectral and dynamical phenomena: a graph whose Laplacian has purely singular continuous spectrum (Theorem 6.1); a partly flat band whose eigenvectors must have infinite support (Theorem 4.1); eigenvalue functions that are continuous but nowhere differentiable (Proposition 5.1); a general ballistic transport formula (Theorem 7.2); sharp polynomial Combes-Thomas decay (Proposition 8.2); and the central counterexample (Theorem 1.1(2)): a graph with purely absolutely continuous spectrum and ballistic transport that nevertheless satisfies no dispersive estimate, giving a negative answer to Open Question 9 of [28] in the non-locally finite setting. The paper also treats the fractional Laplacian, proving a transition from ballistic to super-ballistic motion in dimension one for alpha <= 1/4 (Proposition 7.1).","tokens_in":45860,"tokens_out":22767,"duration_ms":211892,"significance":"The main counterexample is supported by explicit, essentially self-contained computations: the Floquet function b(theta)=|1/2-theta| has summable, nonnegative, symmetric Fourier coefficients, b' is nonzero a.e., and the kernel formula (7.8)-(7.9) yields the uniform lower bound ||e^{-itH}delta_0||_infty > 1/pi for all times. This is a convincing and falsifiable negative answer within the stated class. The transport statements are measured by the Euclidean coordinate operator x rather than graph distance; this is explicitly justified in Remark 7.4, since graph distance degenerates in non-locally finite graphs. That choice is a modeling decision, not a gap. The paper's exact solvability and the breadth of phenomena make it a useful contribution to the spectral theory of essentially local periodic operators, provided the local errors discussed below are corrected.","major_comments":[{"comment":"The claimed asymptotic |(e^{it(-Delta)^alpha}delta_0)(0)| ~ t^{-1/2} for every alpha < 1 is false for 1/2 < alpha < 1. The proof itself shows that the integral over [0, pi/4] is ~ t^{-1/beta} with beta = 2 alpha in (1,2), while the integral over [pi/4, pi/2] is ~ t^{-1/2}. Since 1/beta > 1/2, the endpoint contribution dominates, so the correct statement is ~ t^{-1/(2 alpha)} for 1/2 < alpha < 1, and ~ t^{-1/2} for 0 < alpha <= 1/2 (with alpha = 0 as an exceptional, trivial case). The sentence beginning 'Quite surprisingly, this is not the case' and any conclusions drawn from Lemma 7.12 must be revised accordingly.","section":"§7.2.4, Lemma 7.12"},{"comment":"For h(theta) = (theta - 1/2)^2, Corollary 7.7 gives lim ||x e^{-itH} psi||^2 / t^2 = (1/(4 pi^2)) integral_0^1 (2 theta - 1)^2 |psihat(theta)|^2 dtheta = (1/pi^2) integral_0^1 (theta - 1/2)^2 |psihat(theta)|^2 dtheta, not (1/(2 pi^2)) integral_0^1 |theta - 1/2| |psihat(theta)|^2 dtheta as printed. The displayed formula appears to use the derivative of the function b instead of a. The qualitative conclusion that the limit is nonzero for nontrivial psi remains correct, but the constant and integrand need correction.","section":"Example 7.8(c)"}],"minor_comments":[{"comment":"In the first displayed integral of §7.2.1 the phase is written e^{ita(theta)}, but the surrounding calculation uses the function c(theta) from (3.1c); replace a by c.","section":"§7.2.1, Eq. (7.5)"},{"comment":"The fractional Laplacian statements should explicitly assume 0 < alpha < 1 (or alpha > 0), since alpha = 0 gives (-Delta)^0 = I, for which the super-ballistic claim and the t^{-1/2} decay statement do not hold.","section":"Proposition 7.1 and Lemma 7.12"},{"comment":"Since the discontinuity set J of nu' need not be closed, the proof should either define J to be a closed null set containing the discontinuities or work on the complement of the closure; the hypothesis already mentions the closure, so this is a consistency issue in the wording.","section":"Theorem 7.2"},{"comment":"The statement 'any initial state psi != 0 spreads out at ballistic speed' is proved in Corollary 7.7 only under the condition ||x psi|| < infinity; the theorem should include this hypothesis or state the convention explicitly.","section":"Theorem 1.1(1)"},{"comment":"The displayed derivative formula contains a typographical garbling, '-2 beta beta pi it', which appears to be intended as '-i t 2^beta beta pi'; the divergence argument is unaffected by the constant.","section":"§7.1.1, proof of Proposition 7.1"}],"recommendation":"major_revision","confidential_remarks":"The central counterexample of Theorem 1.1(2) is sound, and the paper's scoping of transport to the Euclidean coordinate operator is explicitly and honestly justified in Remark 7.4. The two mathematical errors I found are local and correctable; they do not affect the main theorem, but they do require revision before the paper can be accepted as is."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Main thing you should know: this paper is worth your time. The headline result—a Z-periodic graph with purely ac spectrum and ballistic transport but no dispersion, the \"sliding tents\" example—is proved by an explicit computation that I checked at least the key step of. The lower bound ||e^{-itH}δ_0||_∞ ≥ 1/π is right. This is a real negative answer to Open Question 9 of Damanik–Malinovitch–Young, scoped to the non-locally-finite periodic setting, and that scope is stated honestly.\n\nWhat's new: the paper shows that dropping local finiteness while keeping summable, symmetric, nonnegative weights changes the spectral/dynamical picture substantially. You get purely singular continuous spectrum for a free Laplacian (Theorem 6.1), flat bands with no compactly supported eigenvectors (Theorem 4.1), continuous nowhere-differentiable Floquet functions (Proposition 5.1), sharp polynomial Combes–Thomas decay (Proposition 8.2), and a super-ballistic phase for the fractional Laplacian (Proposition 7.1). The constructions are explicit and the proofs are mostly self-contained, with the main wave-packet computations done by hand. That's real value.\n\nSoft spots: the transport results all use the Euclidean coordinate operator x (multiplication by k in the Floquet picture) rather than graph distance. The paper flags this in Remark 7.4 and notes that graph distance is degenerate here—d(0,n)≤2 in the odd-weight example—so graph-distance transport would not even measure spreading. That is a modeling choice, not a gap, but it means the dynamical half of the paper should be read as 'transport with respect to the Euclidean coordinate,' not as a claim about the graph metric.\n\nA second caveat: Theorem 7.2 assumes the closure of the jump set of the eigenvalue-counting function has measure zero, and the paper only says it knows no counterexample. That's a minor gap in generality, but it does not affect the concrete ν=1 examples, where the condition is empty.\n\nThe proofs of Theorem 6.1 and some external results rely on standard cited theorems (Anderson–Housworth–Pitt, Olver). That's acceptable.\n\nWho this is for: anyone working on periodic operators, spectral theory of graphs, or quantum walks on lattices. It deserves a serious referee—the central counterexample is important enough that the field needs to know whether it's right, and I believe it is. Strong accept for review; expect some revision on presentation, not on core math.","headline":"A genuinely new, rigorously proved counterexample in the non-locally-finite periodic setting that answers an open question about dispersion; the main caveat is the Euclidean-coordinate definition of transport, which is forced but limits the dynamical claims.","tokens_in":46404,"tokens_out":3983,"would_cite":true,"duration_ms":35591,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81Q10","42A32","42B05"],"pacs":[],"model":"deepseek-v4-flash","headline":"Waves travel at ballistic speed yet never flatten on this crystal","keywords":["periodic graphs","non-locally finite graphs","singular continuous spectrum","ballistic transport","dispersion estimates","fractional Laplacian","flat bands","Floquet theory"],"falsifier":"Directly evaluate the explicit kernel (7.8) for the tent Floquet function $b(\\theta)=|\\tfrac12-\\theta|$: the sup norm of $e^{-itA_{\\Gamma}}\\delta_n$ should remain at least $1/\\pi$ for every $t$, with peaks near $n\\pm k$ at $t\\approx 2\\pi k$. A numerical or analytical check finding any time at which the sup norm decays to zero would refute the no-dispersion claim; conversely, confirming the lower bound at the predicted peak positions verifies the core counterexample.","tokens_in":45485,"feed_emoji":"🌊","tokens_out":6978,"duration_ms":66915,"temperature":0.7,"pith_summary":"This paper studies periodic graphs in which a vertex can have infinitely many neighbours, provided the edge weights are nonnegative, symmetric, and summable. Its central claim is that dropping local finiteness, while keeping this natural summability, unlocks spectral and dynamical phenomena that are impossible for ordinary crystals. The flagship example is a one-dimensional crystal whose adjacency operator has purely absolutely continuous spectrum and ballistic transport but no dispersive estimate: a wave packet splits into two travelling tents whose peaks move at linear speed without decaying. That example answers negatively an open question on whether ballistic motion forces dispersive flattening. The same framework also produces a crystal whose free Laplacian has purely singular continuous spectrum, and a crystal with a flat band whose eigenvectors must have infinite support.","feed_headline":"No-dispersion crystal: ballistic waves keep sharp peaks forever","feed_subtitle":"A periodic graph with inverse-square long hops splits wave packets into travelling tents whose peaks never decay, refuting a proposed…","key_machinery":"The load-bearing mechanism is the Floquet transform for a crystal with one vertex per fundamental cell: it turns the adjacency operator into multiplication by a function $h$ on the torus whose Fourier coefficients are exactly the edge weights. Because the weights are summable, $h$ lives in the Wiener algebra, and the paper works backwards, choosing simple geometric symbols — the parabola $(\\theta-\\tfrac12)^2$, the tent $|\\tfrac12-\\theta|$, and a piecewise-linear function with a flat middle interval — and then reading off spectra and dynamics from oscillatory integrals of $e^{ith(\\theta)}$. The no-dispersion claim is carried by an explicit formula for the evolution kernel of the tent symbol, which shows two travelling peaks whose heights stay bounded below.","core_discovery":"On the paper's own terms, the central discovery is that periodic graphs with summable, non-negative, symmetric weights — a subclass of essentially local operators — support spectral and dynamical behaviour that locally finite crystals forbid. Theorem 1.1(2) constructs the $\\mathbb{Z}$-periodic graph with $w(k)=k^{-2}$ for odd $k$ and $w(k)=0$ for even $k$: its adjacency operator is unitarily equivalent to multiplication by the tent function $|\\tfrac12-\\theta|$ on the torus, which gives purely absolutely continuous spectrum and ballistic transport for every non-zero initial state. Yet the evolution of a delta state violates dispersion: $\\|e^{-itA_{\\Gamma}}\\delta_n\\|_\\infty > c > 0$ for all times, with the wave splitting into two tents whose peaks travel at linear speed and decay polynomially away from the peaks. The paper reads this as a negative answer to an open question on whether ballistic motion forces dispersive flattening, and it accompanies the counterexample with other exotic phenomena: purely singular continuous spectrum for a free Laplacian, a partly flat band with no compactly supported eigenvectors, and a fractional-Laplacian phase transition at $\\alpha=\\tfrac14$.","pith_inferences":["Editorial inference: because many vertices are neighbours of the origin, graph-distance spreading is degenerate on these graphs; the paper's ballistic and super-ballistic statements are about Euclidean coordinate spreading, so a reader interested in graph-geometric transport should ask which conclusions survive that change of observable.","Editorial inference: the ease of tuning the Floquet symbol suggests a testable numerical programme — random or periodic zero sets in the $k^{-2}$ weights, as sketched in the paper's final problems — could be probed for their spectral type and transport exponents.","Editorial inference: if the open question on dispersion was intended for all essentially local periodic Hamiltonians, the sliding-tents example suggests the question can only survive under extra regularity or decay assumptions on the weights."],"forward_implications":["If the central claim is correct, the locally finite dichotomy — purely absolutely continuous spectrum with at most finitely many flat bands and no singular continuous spectrum — does not extend to summable-weight periodic graphs.","An affirmative answer to the dispersive-estimate question that had been left open is ruled out for this class: ballistic motion and non-decaying peaks can coexist.","Flat bands in this setting need not come with compactly supported eigenvectors, so any characterization of flat bands for non-locally finite crystals has to account for infinite-support modes.","For the fractional Laplacian on $\\mathbb{Z}$, transport is ballistic for $\\alpha>\\tfrac14$ and super-ballistic for $\\alpha\\le\\tfrac14$, with no analogous transition in higher dimension.","Combes-Thomas decay for these graphs is only polynomial in general, with the sharp rate $n^{-2}$ for the model Floquet functions."],"supporting_citations":[{"why":"poses the open question on ballistic transport and dispersion that Theorem 1.1(2) answers negatively.","marker":"[28]"},{"why":"supplies the Floquet-based ballistic transport framework and the proof template for Theorem 7.2.","marker":"[13]"},{"why":"records the absence of singular continuous spectrum for locally finite periodic graphs, the contrast for Theorem 6.1.","marker":"[45]"},{"why":"establishes the locally finite flat-band facts, compact-support eigenvectors and analyticity, that Theorem 4.1 shows fail here.","marker":"[77]"},{"why":"gives the spectral theorem for multiplication operators by nondifferentiable functions, the source of the purely singular continuous example.","marker":"[8]"},{"why":"provides the Zygmund-class result placing the example function with the non-differentiability needed in Theorem 6.1.","marker":"[88]"}],"fun_headline_variants":["Ballistic waves that refuse to flatten in a periodic graph","No dispersal: periodic graph keeps wave peaks sharp","Sharp tent waves travel without spreading in crystal","Ballistic motion without dispersion in nonlocal crystal","Wave peaks never decay in a periodic graph"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"All dynamical conclusions measure spreading by the Euclidean coordinate $k$ in the fundamental-cell representation, not by graph distance, because on these graphs most vertices are neighbours of the origin; if one insisted on graph-distance transport, the notions of ballistic, super-ballistic, and no-dispersion would have to be redefined.","fun_headline_variants_meta":{"raw":{"variants":["Ballistic waves that refuse to flatten in a periodic graph","No dispersal: periodic graph keeps wave peaks sharp","Sharp tent waves travel without spreading in crystal","Ballistic motion without dispersion in nonlocal crystal","Wave peaks never decay in a periodic graph"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000953,"raw_usage":{"total_tokens":4080,"prompt_tokens":973,"completion_tokens":3107,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":589,"completion_tokens_details":{"reasoning_tokens":3036}},"tokens_in":589,"tokens_out":3107,"duration_ms":24553,"temperature":1.0,"reasoning_tokens":3036,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:40:20.829605+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly evaluate the explicit kernel (7.8) for the tent Floquet function $b(\\theta)=|\\tfrac12-\\theta|$: the sup norm of $e^{-itA_{\\Gamma}}\\delta_n$ should remain at least $1/\\pi$ for every $t$, with peaks near $n\\pm k$ at $t\\approx 2\\pi k$. A numerical or analytical check finding any time at which the sup norm decays to zero would refute the no-dispersion claim; conversely, confirming the lower bound at the predicted peak positions verifies the core counterexample.","supporting_citations":[{"cited_title":"Damanik, T","cited_arxiv_id":null,"evidence_quote":"poses the open question on ballistic transport and dispersion that Theorem 1.1(2) answers negatively."},{"cited_title":"From Complex Analysis to Operator Theory: A Panorama, In Memory o f Sergey Naboko","cited_arxiv_id":null,"evidence_quote":"supplies the Floquet-based ballistic transport framework and the proof template for Theorem 7.2."},{"cited_title":"Higuchi, Y","cited_arxiv_id":null,"evidence_quote":"records the absence of singular continuous spectrum for locally finite periodic graphs, the contrast for Theorem 6.1."},{"cited_title":"Sabri and P","cited_arxiv_id":null,"evidence_quote":"establishes the locally finite flat-band facts, compact-support eigenvectors and analyticity, that Theorem 4.1 shows fail here."},{"cited_title":"Zygmund, Trigonometric Series, Third Edition, Volumes I&II combined, CUP 2002","cited_arxiv_id":null,"evidence_quote":"provides the Zygmund-class result placing the example function with the non-differentiability needed in Theorem 6.1."}],"review_version":1}