{"id":"9526e538-cc6f-40fe-9803-9bb09e62a84b","arxiv_id":"2607.16958","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A sufficiently strong ferromagnetic interaction forces all zeros of the Blume-Capel partition function onto the imaginary axis even when the single-spin part violates the Lee-Yang condition.","lead":"The paper proves that adding ferromagnetic coupling to the Blume-Capel spin-1 model can restore the Lee-Yang property—all partition-function zeros on the unit circle—even when the single-spin partition function fails it. This gives a new sufficient condition on temperature, coupling, and crystal field for a standard statistical-mechanics model.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the dimer-split proof and two-spin stability calculation are internally sound under the stated assumptions.","rationale":"The reader's weakest_assumption identified condition (c) as the key limitation. I agree that this is the main restriction on applicability, but it is explicit, not hidden, and it does not undermine the theorem's internal validity. I independently rechecked the algebraic steps: the trace calculation, the use of θ^2=cosh κ, the factorization of Ψ, the positivity of ω_±, and the no-zero contradiction are all correct. The Lieb-Sokal step is justified because K'' is nonnegative. The typos in the text are unfortunate but do not change the mathematics. Therefore the reader's ACCEPT verdict is appropriate and no verdict adjustment is needed.","tokens_in":10967,"tokens_out":31389,"duration_ms":315239,"concrete_test":"Use a computer algebra system to evaluate the two-spin function Ψ(x_i,x_j) from equation (20) for θ=2, κ=arccosh 4, and perform a dense randomized search over a rectangle in the right half-plane, e.g., Re x ∈ [0.1,5], Im x ∈ [-20,20] for both variables, checking for zeros. If any zero is found, the stability proof fails; alternatively, solve the diagonal case Ψ(h,h)=0 and verify all roots have Re h = 0.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After checking the argument in Section 2, the central stability proof holds. The trace identity (19) correctly represents e^{κ D_i D_j} Z_1(x_i)Z_1(x_j); equation (20) follows from the identity with cosh κ = θ^2 and the factorization with ω_±; the bounds 0<ω_-<ω_+<1 are proven correctly. The contradiction argument using (25)-(26) excludes zeros for Re x_i>0, Re x_j>0, so Ψ is stable. The subsequent application of Lieb-Sokal to add K'' is valid because K'' has nonnegative entries. The only real restriction is condition (c): the perfect matching with a uniform lower bound κ. This is explicit and honestly acknowledged; it excludes Curie-Weiss interactions and forces even-length parallelepipeds, but the claim is not overstated. The text contains minor typos—the inline definition θ=e^{βΔ/2} should be e^{βΔ}/2, and for θ>1 one of the zeros in (7) lies inside the unit circle and one outside—but neither affects the proof or the applications.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses the Lee-Yang property for the spin-1 Blume-Capel model when the single-spin partition function fails to have it (βΔ > ln 2). The main result (Section 2, Statement) is: for N = 2M, if Φ = ∏ Z_1(x_i) and the symmetric ferromagnetic interaction matrix K satisfies K_{ij} ≥ 0 and admits a perfect matching with uniform lower bound K_{i_k j_k} ≥ κ > 0, then F = exp(½∑ K_{ij} D_i D_j) Φ is stable whenever e^{2βΔ}/4 = θ² ≤ cosh κ. Consequently, for the nearest-neighbor Blume-Capel model on Z^d, finite-volume partition functions on even-length parallelepipeds have purely imaginary zeros in h whenever e^{2βΔ}/4 ≤ cosh(βJ). The proof splits K into a dimer part K′ and a remainder K″, computes the two-spin trace exactly, proves the stability of the dimer factor by an explicit hyperbolic-function argument, and then uses the Lieb-Sokal theorem to add K″.","tokens_in":11239,"tokens_out":19054,"duration_ms":176968,"significance":"If correct, the result is a first rigorous example in which ferromagnetic interactions restore the Lee-Yang property under conditions where the single-spin condition fails. The proof is transparent and mostly elementary; its key ingredients are the dimer decomposition and the two-spin stability calculation. The theorem is a sufficient condition, and the limitations (perfect matchability, uniform lower bound, even number of sites) are stated explicitly and honestly. The condition (8) is falsifiable and the authors compare it with known one-dimensional numerical data. The paper is likely to be of interest to mathematical physicists working on Lee-Yang theory and the Blume-Capel model.","major_comments":[],"minor_comments":[{"comment":"The displayed chain 1+e^{-βΔ}(z+z^{-1}) = z^{-1}(z^2+2θz+1) is not an identity. With θ=e^{βΔ}/2, the correct relation is Ξ_1(z)=e^{-βΔ}z^{-1}(z^2+2θz+1). The zeros in (7) are correct, and the error does not affect the proof, but it should be corrected.","section":"Eq. (6) and definition of θ"},{"comment":"The parameter θ is defined inconsistently: Eq. (6) and the text after Eq. (19) define θ=e^{βΔ/2}, whereas Eq. (12) and the trace calculation (20) require θ=e^{βΔ}/2. Please make the definition uniform.","section":"Eqs. (6), (12), (20)"},{"comment":"For θ>1 it is stated that both zeros z_± are real and lie outside the unit circle. In fact, one zero lies inside and one outside. The needed conclusion is only that they are not on the unit circle, so the statement should be rephrased.","section":"after Eq. (7)"},{"comment":"The sentence 'pick positive κ ≤ κ that verifies' uses the same symbol κ for the lower bound from (11) and for the chosen value arccosh θ². Use distinct notation and state explicitly that the chosen value must not exceed the given uniform bound.","section":"Section 2, proof, Eq. (13)"},{"comment":"The Statement itself does not explicitly mention symmetry of K or of F, although symmetry is needed to pass from stability to purely imaginary zeros of Z_N(h)=F(h,...,h). This should be stated or remarked upon for clarity.","section":"Statement and Applications"}],"recommendation":"minor_revision","confidential_remarks":"The result appears correct and the limitations are honestly stated. The paper is a good fit for the journal; after the typographical issues are fixed, I would support acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper proves that for the Blume-Capel model with sufficiently strong ferromagnetic coupling, the partition function zeros lie on the imaginary axis even when the single-spin partition function fails the Lee-Yang condition. The mechanism is a dimer decomposition: pair the sites, put a lower bound κ on the interaction within each pair, and compute the two-spin trace exactly. The stability calculation checks out; the contradiction argument with ε± is valid. Then the Lieb-Sokal theorem lifts stability to the full interaction matrix. The sufficient condition e^{2βΔ}/4 ≤ cosh(βJ) is explicit and checkable, and it genuinely goes beyond the direct Lieb-Sokal implication. I agree with the reader that this is the first rigorous instance of interaction-induced Lee-Yang property for this model; the numerical prior [13] is cited honestly.\n\nThe weakest part is the matching condition (c): the interaction matrix must admit a perfect matching with a uniform positive lower bound. That excludes the Curie-Weiss interaction and forces even-length parallelepipeds for lattice applications. But the authors state this plainly and do not overclaim. The thermodynamic-limit discussion is careful: they note the bound must be uniform along the sequence.\n\nSoft spots are mostly typographical. Eq. (6) as written is false (coefficient mismatch), though the zeros in (7) are correct. The symbol κ is used both for the lower bound and the chosen arccosh value. These are easy fixes. The paper also notes the result is sufficient and likely not sharp, and includes a numerical example where the bound is exceeded.\n\nFor whom: mathematical physicists working on Lee-Yang theory or partition function zeros. The paper deserves a serious referee; it is not a breakthrough but a solid, self-contained result with a clear proof. I would accept after minor revision.","headline":"A sound, modest extension of Lee-Yang theory to the Blume-Capel model with strong anisotropy, based on a dimer-decomposition trick; worth publishing after small fixes.","tokens_in":11667,"tokens_out":2714,"would_cite":true,"duration_ms":27206,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["82B20","82B26","30C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"Ferromagnetic coupling can force the Blume-Capel partition function's zeros onto the imaginary axis.","keywords":["Blume-Capel model","Lee-Yang theorem","partition function zeros","stable entire functions","Lieb-Sokal theorem","ferromagnetic interaction","single-ion anisotropy","dimer decomposition"],"falsifier":"Take a four-site periodic chain with nearest-neighbor coupling J and anisotropy Δ satisfying e^{2βΔ}/4 ≤ cosh(βJ); the theorem asserts all zeros of Z_4 are purely imaginary. A single computed zero off the imaginary axis in that parameter range would refute the statement.","tokens_in":10887,"feed_emoji":"🧲","tokens_out":6737,"duration_ms":65119,"temperature":0.7,"pith_summary":"The paper aims to extend the Lee-Yang theorem to the Blume-Capel model in a regime where the standard single-spin criterion fails. In the Blume-Capel model, the single-ion anisotropy term makes the single-spin partition function lose the Lee-Yang property once βΔ exceeds ln 2, so the classical Lieb-Sokal theorem can no longer guarantee the property for the full partition function. The authors prove that a sufficiently strong ferromagnetic interaction can overcome this failure: for any even set of spins that can be paired into dimers with coupling at least κ, all zeros of the full partition function are purely imaginary whenever e^{2βΔ}/4 ≤ cosh κ. This is the first rigorous demonstration that ferromagnetic interaction can induce the Lee-Yang property. If correct, it makes the full machinery of Lee-Yang theory available for the Blume-Capel model in that parameter range.","feed_headline":"Ferromagnetism can rescue the Lee-Yang property for Blume-Capel","feed_subtitle":"First rigorous demonstration that ferromagnetic bonds can induce the property even as the single-spin criterion fails.","key_machinery":"The proof pivots on a dimer decomposition of the interaction matrix. The matrix K is split into a dimer part K' that carries exactly the uniform coupling κ along a chosen perfect matching, and a residual nonnegative part K''. One first proves stability for the factorized dimer function G = ∏ Ψ(x_{i_k}, x_{j_k}), where Ψ is the two-spin partition function of a dimer, and then applies the Lieb-Sokal theorem to the residual interaction to lift that stability to F. The key identity reduces Ψ to a factorized trigonometric form: Ψ = 4 e^{-2βΔ + κ} (c_+ + ω_+ c_-)(c_+ + ω_- c_-), with c_± = cosh((x_i ± x_j)/2) and ω_± = e^{-κ}(θ ± √(θ²−1)). The condition θ² ≤ cosh κ ensures 0 < ω_- < ω_+ < 1, which","core_discovery":"The paper's central result is a stability theorem for the differential-operator action on single-spin partition functions. Let N = 2M, let Φ be the product of N single-spin partition functions Z_1(x) = 1 + 2 e^{-βΔ} cosh x, and let K be a symmetric nonnegative matrix whose entries satisfy K_{i_k,j_k} ≥ κ > 0 for a partition of the indices into M disjoint pairs. Then the function F(x) = exp(½ Σ K_{ij} D_i D_j) Φ is stable — meaning it does not vanish when all arguments have positive real part — provided e^{2βΔ}/4 = θ² ≤ cosh κ. Stability of this symmetric function implies that the Blume-Capel partition function Z_N(h) = F(h,...,h) has only purely imaginary zeros. For the nearest-neighbor mode","pith_inferences":["One might expect the dimer condition to be replaceable by a more global condition: the proof uses only the two-spin stability of a single strongly coupled pair, so any decomposition into small stable clusters of fixed size should work with an appropriate cluster stability lemma.","The same mechanism could apply to other spin models where the single-spin partition function loses the Lee-Yang property but the interaction can compensate, such as higher-spin Ising models with single-ion anisotropy.","The theorem suggests a possible boundary effect: for fixed βΔ > ln 2, there may be a sharp critical J_c(Δ) above which the Lee-Yang property holds; the current bound gives a sufficient J, and numerical evidence indicates the true threshold is close.","A practical test would be to compute the zero set of the finite-volume partition function for small systems at parameters just above the proven curve, which could either support the conjecture that the bound is near-sharp or reveal structure the dimer argument misses."],"forward_implications":["For the nearest-neighbor Blume-Capel model on Z^d, the partition function on any parallelepiped with an even number of sites in one direction has purely imaginary zeros whenever e^{2βΔ}/4 ≤ cosh(βJ).","The limiting free energy is analytic in the complex magnetic field for Re h ≠ 0, provided the thermodynamic limit is taken along such even parallelepipeds with a uniform lower bound on the coupling.","The same construction applies to long-range interactions by taking κ = β times the largest coupling in the direction of a matched chain.","The result gives a sufficient condition; violating it does not imply loss of the Lee-Yang property, and numerical examples in one dimension suggest the true boundary may lie above the proven curve."],"fun_headline_variants":["Ferromagnetic bonds restore Lee-Yang property in Blume-Capel","Blume-Capel: Ferromagnetism induces Lee-Yang zeros","New proof: ferromagnetism triggers Lee-Yang property","Ferromagnetic coupling revives Lee-Yang for Blume-Capel"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The argument depends on the existence of a perfect matching of the spin set into pairs whose ferromagnetic coupling is uniformly bounded below by a positive κ; if no such dimer decomposition exists, the proof gives nothing.","fun_headline_variants_meta":{"raw":{"variants":["Ferromagnetic bonds restore Lee-Yang property in Blume-Capel","Blume-Capel: Ferromagnetism induces Lee-Yang zeros","New proof: ferromagnetism triggers Lee-Yang property","Ferromagnetic coupling revives Lee-Yang for Blume-Capel"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000586,"raw_usage":{"total_tokens":2609,"prompt_tokens":781,"completion_tokens":1828,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":525,"completion_tokens_details":{"reasoning_tokens":1752}},"tokens_in":525,"tokens_out":1828,"duration_ms":13043,"temperature":1.0,"reasoning_tokens":1752,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T19:26:24.244714+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a four-site periodic chain with nearest-neighbor coupling J and anisotropy Δ satisfying e^{2βΔ}/4 ≤ cosh(βJ); the theorem asserts all zeros of Z_4 are purely imaginary. A single computed zero off the imaginary axis in that parameter range would refute the statement.","supporting_citations":[],"review_version":1}