{"id":"a14ffe0e-06a8-4b7b-9651-85be1e9e4a84","arxiv_id":"2608.08669","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"For the Lee-Yang series of minimal Liouville gravity, the four Frobenius and spectral-curve descriptions agree at genus zero after one per-insertion factor, and the resonance map is a tree-level Kontsevich frame change.","lead":"This paper maps four equivalent mathematical formulations of a simple model of two-dimensional quantum gravity and shows that their three-point predictions match the known conformal field theory values. The result clarifies how the 'resonance' change of variables used in some formulations is really a change of coordinates produced by the Kontsevich matrix model transform.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The four-corner amplitude agreement rests on the inherited universality projection; at (2,7) the filter, not κ, already removes three of four forbidden triples, so the dictionary is underdetermined without a higher-p check.","rationale":"The reader identified the universality projection as the weakest assumption, and I agree that it is the most load-bearing element of the argument. The strongest claim has two parts: normalisation-independent universal ratios, and full amplitude-level agreement including the signed Verlinde matrices. The ratio part is a genuine check, but it only uses fusion-allowed triples and is insensitive to the projection, since the projected-out entries are not inserted into the ratio. The amplitude-level part, however, requires every fusion-forbidden amplitude to vanish, and in the explicit SC-y computations these vanish mostly because of the projection, not because of the resonance-plus-κ mechanism. At (2,5) the filter alone removes the one forbidden entry; at (2,7) the filter removes three of the four forbidden entries and κ is fixed by the single remaining one. Consequently, the claim that a single per-insertion factor controls the whole tower is weaker than stated: it is an over-determined check only at the level of the surviving amplitudes, not of the forbidden sector. This does not make the paper wrong; it makes the amplitude-level four-corner dictionary conditional on a convention that is not independently derived here. The other gaps noted by the reader, such as the unproved FM-x↔SC-x link and the off-background FM-y↔SC-y equivalence, are scope limitations explicitly acknowledged in Section 4.3 and do not by themselves undermine the correctness of what is computed. The decisive test is therefore a full (2,11) amplitude-level computation, where the resonance is already derived in Appendix D but no amplitude comparison is currently presented. If that test passes, the conditional verdict can be upgraded; if it fails, the four-corner amplitude dictionary is not correct as stated.","tokens_in":37265,"tokens_out":13645,"duration_ms":148855,"concrete_test":"Compute the full (2,11) SC-y three-point amplitudes from the Chebyshev-curve topological recursion, apply the resonance map (3.11), the per-insertion factor κ=-2, and the universality projection (B.36), and compare every fusion-forbidden triple with zero and every fusion-allowed triple with the signed Verlinde entries of (4.7). In particular, list all forbidden triples whose u0 power is odd or negative, since those survive the projection and must be cancelled by the resonance and κ: if any of them fails to cancel, the universality-plus-κ mechanism is not the correct dictionary; if all of them cancel, the projection is supported beyond the two worked points (2,5) and (2,7).","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that all four corners compute the same dispersionless tau-function at the MLG background after the resonance map on the y-side corners and a single per-insertion factor κ=-2 on the spectral-curve corners. The amplitude-level agreement is not actually driven by that factor alone. At (2,5), the forbidden (1,1,2) amplitude is removed by the universality projection alone, independently of κ. At (2,7), the forbidden (1,1,2), (1,2,3) and (2,2,2) amplitudes are likewise discarded by their u0-power via eq. (B.36), leaving only the (1,1,3) amplitude to exercise the κ cancellation through A0_3(1,1,3)∝(2+κ). Thus the 'single per-insertion factor' is tested by exactly one surviving forbidden amplitude; all other fusion-forbidden zeros are put in by the inherited filter described in Section 2.1 and Appendix B. Since the paper does not derive this projection from an independent physical principle, but explicitly inherits it from [12,15], the equality of the four corners at amplitude level is conditional on the projection being the correct dictionary. The universal-ratio agreement on allowed triples is a genuine normalisation-independent check, but it does not test the projection, because the discarded entries are absent from the ratio by construction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the (2,p) Lee-Yang minimal Liouville gravity for p=5,7,9,11 and claims that four algebraic descriptions—FM-y (A1 Frobenius manifold), FM-x (A_{p-1} Frobenius manifold), SC-y (standard Chebyshev spectral curve with resonance transformation), and SC-x (swapped Chebyshev curve without resonance)—produce the same genus-zero dispersionless tau-function at the minimal-Liouville-gravity background, after the resonance map on the y-side corners and a single per-insertion factor kappa=-2 on the spectral-curve corners. The paper computes two-point and three-point amplitudes and universal three-point ratios explicitly for (2,5) and (2,7), reproduces the signed Verlinde matrices, proves a defining tau-relation at the background for all (2,p), and derives the resonance coefficients from the spectral curve through a genus-zero Kontsevich-type frame change. It also reconciles Artemev's compact resonance formula with the Belavin-Dubrovin-Mukhametzhanov Jacobi-polynomial construction.","tokens_in":37564,"tokens_out":5803,"duration_ms":63615,"significance":"If the claims hold, the paper provides a useful explicit dictionary among four currently used formulations of (2,p) minimal Liouville gravity. The normalisation-independent three-point ratios are a genuinely scheme-free test and agree with CFT in all four corners; Proposition 4.1 is a clean structural statement at the background; and the reproduction of the signed Verlinde matrices, including the non-unitary signs, is a nontrivial overdetermined check. The explicit computations in Appendices B and C are detailed and reproducible, and the reconciliation of the two resonance conventions in Appendix A is a concrete contribution. The paper is honest about its limitations, explicitly stating that the four-point level and bare correlators are not addressed and that FM-x to SC-x remains an open problem.","major_comments":[{"comment":"The universality projection is load-bearing for the amplitude-level equivalence, but it is inherited from Refs. [12,15] and not derived from an independent physical principle. At (2,5) the fusion-forbidden amplitude A0_3(1,1,2) is removed solely by the projection because it sits at the discarded u0^0 power; at (2,7) three of the four forbidden triples, namely (1,1,2), (1,2,3) and (2,2,2), are likewise discarded by Eq. (B.36), leaving only the (1,1,3) entry to exercise the kappa cancellation of Eq. (C.19). The universal-ratio agreement on fusion-allowed triples does not test this projection, since the discarded entries are absent from the ratios by construction. Given that the paper claims validation across (2,5)-(2,11) while the amplitude-level checks are performed only for (2,5) and (2,7), the four-corner amplitude agreement is conditional on the universality projection being the correct dictionary. The authors should either derive the projection from the worldsheet theory, or verify the amplitude-level cancellation on at least (2,9) and (2,11), where more forbidden triples survive the filter and would genuinely constrain kappa.","section":"§2.1 and Eq. (B.36)"},{"comment":"The derivation of the resonance from the curve is partly circular. In Appendix D the normalization s_m = -2 tau_m is fixed by 'demanding that the linear part reproduce (3.11), t_k = p tau_k', but Eq. (3.11) is precisely the target map that the derivation is supposed to produce. Consequently the subsequent reproduction of the quadratic and mixed resonance coefficients, including tau_2^2 -> t_5 at (2,11), is not an independent prediction: the only genuinely input-free datum is the cosmological coefficient in Eq. (D.3), which follows from the background times. To make the 'derivation from the curve' claim load-bearing, the linear normalization must be fixed by an independent argument, for example from the background values in Table D.2 alone or from the definition of the Liouville couplings, rather than by imposing the target formula.","section":"Appendix D, Eq. (D.10)"},{"comment":"The generalised Kontsevich transform with potential Phi^p/p is introduced as the kernel that maps the y-side tau-function to the x-side tau-function, but no derivation of this kernel from the Bertola-Eynard-Harnad two-matrix model or from the spectral curve is provided; the saddle-point analysis in Eqs. (5.4)-(5.6) only shows that the transform reduces to a shift of the KdV times. The paper itself states in §5.3 that on the physical sector the conjectural Kontsevich kernel 'degenerates ... from an integral transform to this mere change of variables', so the matrix-model 'derivation' of the resonance is better described as a consistency check of the frame-change interpretation, not a derivation of the resonance from first principles. The claim in the Summary of Results that the resonance is derived from the Kontsevich transform should be softened or justified by an independent check that the integral kernel, rather than just its classical limit, is the correct map between the two tau-functions.","section":"§5.2 and Eq. (5.2)"}],"minor_comments":[{"comment":"There are typographical spacing errors in the title and abstract: 'for(2, p)minimal' and similar missing spaces should be corrected.","section":"Title and abstract"},{"comment":"The text 'therefore both SC-yand SC-zxhave natural integrability interpretations' contains the typo 'SC-zx'; it should read 'SC-y and SC-x'.","section":"§3.2, paragraph after Eq. (3.20)"},{"comment":"The text says the universal ratio will be fixed by 'a single universal normalisation that will be fixed in Section 4', but the normalisation kappa is actually fixed in Section 3.1 by Eq. (3.10). The cross-reference should be updated.","section":"§4.4, Eq. (4.10)"},{"comment":"In the coefficient table, the entry for lambda_0^3 is marked 'N/A' because the BDM cubic block is missing, but the following paragraph derives the value 8/27951 indirectly. The table would be clearer if this derived value were displayed with a footnote explaining that it comes from the Artemev side combined with Eq. (A.12).","section":"Appendix A, table after Eq. (A.14)"},{"comment":"Item 2 says 'The full amplitude-level dictionary, with every normalisation fixed', but the paper only computes two- and three-point amplitudes on the sphere; the four-point level is explicitly deferred in the same section. The wording 'full amplitude-level dictionary' should be qualified to 'full three-point dictionary at genus zero' to avoid overstating the scope.","section":"§6, Summary of Results"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid computational contribution to the (2,p) minimal string dictionary, but the two load-bearing issues above—the inherited universality projection and the circular normalization in Appendix D—need to be addressed before publication. The universal-ratio agreement is the strongest independent evidence, but it does not by itself establish the amplitude-level four-corner identity. The authors should either provide the higher-p amplitude checks or frame the main claim as conditional on the universality projection. The four-point discrepancy already noted in Ref. [17] also limits the 'equivalence' language used in the abstract and introduction."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a serious, mostly careful explicit check of the four-corner dictionary for (2,p) Lee-Yang MLG at genus zero. What is actually new: the normalisation-independent three-point ratio comparison across all four corners for (2,5) and (2,7), Proposition 4.1 establishing the defining tau relation at the MLG background for all (2,p), the explicit reconciliation of Artemev's compact resonance formula with the BDM Jacobi-polynomial construction, and the genus-zero derivation of resonance coefficients from the curve via the Kontsevich transform, including genuinely mixed coefficients at (2,11). The appendices are detailed, internally consistent, and reproducible; the ratio agreement is scheme-free and checks against CFT. The paper is also honest about what remains open: FM-x to SC-x is explicitly unverified, off-background FM-y to SC-y equivalence is unproved, and the four-point discrepancy of [17] is acknowledged.\n\nThe main structural weakness is the universality projection. It is inherited from [12,15] and not derived from an independent physical principle. At (2,5) the forbidden (1,1,2) amplitude is removed by the filter alone; at (2,7), three of the four forbidden triples are discarded by that same projection, leaving only (1,1,3) to exercise the cancellation that fixes kappa=-2. So the per-insertion factor is tested by exactly one surviving forbidden amplitude, and the universal ratios on allowed triples do not test the filter at all, since the discarded entries are absent from the ratios by construction. The resonance derivation in Appendix D also fixes the deformation normalisation s_m=-2 tau_m by requiring the linear part to match (3.11), so it is not fully independent of the target formula. The abstract overstates the completeness: full amplitude-level comparisons are only shown for (2,5) and (2,7), while (2,9) and (2,11) are checked mainly for the resonance coefficients.\n\nThese are addressable issues, not fatal flaws. The explicit computations are real and the paper is a useful reference for anyone working on minimal string dualities. I would send it to a serious referee, with the request that the universality filter be either derived or clearly advertised as a conventional input, and that kappa be tested on at least one more model beyond (2,7). A referee who pushes on this will get a better paper.","headline":"A genuinely useful explicit computation that makes a real dent in the four-corner dictionary, but the amplitude-level agreement leans on an inherited universality filter that is not derived, so the dictionary is less complete than the abstract suggests.","tokens_in":38074,"tokens_out":2341,"would_cite":true,"duration_ms":25514,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The four algebraic formulations of (2,p) minimal Liouville gravity compute the same genus-zero free energy after one resonance map and one per-insertion factor.","keywords":["minimal Liouville gravity","Lee-Yang series","Frobenius manifold","topological recursion","resonance transformation","Chebyshev spectral curve","generalised Kontsevich transform","Verlinde fusion"],"falsifier":"Compute the universal three-point ratio $R(1,1,1)$ for the (2,13) model in all four corners with $\\kappa=-2$; the claim predicts the CFT value $121/195$, and any corner that deviates from it disproves the equivalence. A second check is the fusion-forbidden triple $(1,1,3)$, which must vanish in all four corners; a non-zero value at (2,13) would mark the boundary of the dictionary.","tokens_in":37032,"feed_emoji":"🧮","tokens_out":7153,"duration_ms":69029,"temperature":0.7,"pith_summary":"This paper claims that four known algebraic descriptions of (2,p) minimal Liouville gravity — two Frobenius-manifold formulations and two spectral-curve topological-recursion formulations — compute the same genus-zero free energy at the Lee-Yang background. The universal, normalisation-independent three-point ratios agree with conformal field theory in all four corners, and matching the full amplitudes fixes a single per-insertion factor, $\\kappa=-2$. If this is right, the resonance transformation required on the $y$-side is simply the tree-level frame change of the generalised Kontsevich transform, and the $x\\leftrightarrow y$ swap between the two spectral curves is realised at genus zero as that same change of variables. The claim matters because it reduces a web of independently motivated constructions to one dictionary: one nonlinear map plus one number, verified explicitly across the (2,5), (2,7), (2,9), and (2,11) models.","feed_headline":"Four dual descriptions of (2,p) minimal gravity agree at tree level","feed_subtitle":"Universal three-point ratios match conformal field theory; a single factor κ=-2 fixes the whole dictionary.","key_machinery":"The central object is the genus-zero free energy $F_0=\\log\\tau$ of the (2,p) minimal Liouville gravity at the Lee-Yang background, together with the four-corner diagram relating two Frobenius-manifold descriptions and two spectral-curve descriptions. Three load-bearing pieces carry the argument: the residue prescription that extracts KdV-time derivatives from topological-recursion differentials; the compact resonance map $t_k(\\tau)$ of equation (3.11), whose closed-form sum is a fractional power and hence a Puiseux expansion of the $x\\leftrightarrow y$ swap; and the saddle-point elimination in the generalised Kontsevich integral transform, where eliminating the external field $\\Lambda$ between the two fractional-power moment frames $\\Lambda^{-k/(p-1)}$ and $\\Lambda^{-m/p}$ produces the resonance coefficients as ratios of curve residues. The per-insertion factor $\\kappa=-2$ is fixed once by the vanishing of a fusion-forbidden three-point amplitude and then reproduces every signed Verlinde entry.","core_discovery":"On the paper's own terms, the discovery is that the four established algebraic descriptions of the (2,p) Lee-Yang minimal string — the two Frobenius-manifold constructions on the $y$-side and $x$-side, and the two Chebyshev spectral-curve constructions with and without the $x\\leftrightarrow y$ swap — describe one and the same genus-zero dispersionless tau-function at the MLG background. After the resonance map is applied on the $y$-side corners and the single per-insertion factor $\\kappa=-2$ on the spectral-curve corners, the normalisation-independent three-point ratios coincide with the CFT prediction exactly, and the full amplitudes reproduce the signed Verlinde fusion matrices, negative entries included. The paper further establishes the defining tau relation $u_1^*=2\\,\\partial^2 F_0/\\partial t_s^2$ at the background for every $s$, and realises the $x\\leftrightarrow y$ swap at genus zero as the saddle-point reduction of the generalised Kontsevich transform with potential $\\Phi^p/p$: eliminating the external field between the two fractional-power frames yields the resonance coefficients directly from the curve, with no worldsheet input.","pith_inferences":["If the genus-zero frame change is the dispersionless limit of the Kontsevich transform, then at higher genus the same transform should produce the resonance coefficients order by order in the genus expansion; the first genus-one three-point amplitude is a concrete place to test this.","The universality projection may itself be derivable from the requirement that the tau-function reproduce the CFT selection rules without tuning; if so, the filter and the resonance map would be two aspects of a single normalisation principle rather than two separate conventions.","The mixed resonance coefficient at (2,11) is a sharp prediction that can be checked independently by a direct Liouville CFT three-point computation or by an alternative matrix-model expansion, since it was not fixed by bare residue-time data.","Because the paper leaves four-point amplitudes aside and notes a known four-point discrepancy on one leg, the dictionary is best understood as a statement about universal three-point data; testing four-point universal ratios would map the precise boundary of the equivalence."],"forward_implications":["The three-point ratio $R$ is scheme-free: every operator normalisation cancels, so the equality across all four corners is a direct, tuning-free test of the dualities.","A single per-insertion constant $\\kappa=-2$ converts spectral-curve amplitudes to minimal-gravity normalisation at every multiplicity, and with it topological recursion reproduces the signed Verlinde fusion matrices, including negative entries.","The resonance transformation is not an extra input at genus zero: its coefficients, including the genuinely mixed ones such as $\\tau_2^2\\to t_5=\\tfrac12$ at (2,11), are read off the Chebyshev curve through the Kontsevich frame change.","The defining tau relation $u_1=2\\,\\partial^2F_0/\\partial t_s^2$ holds at the MLG background for every $(2,p)$, so the Frobenius and spectral-curve constructions compute the same tau-function at that point.","The remaining open leg, FM-x versus SC-x, is identified as a conjectural linear identification; if established, the whole four-corner diagram would close without any further normalisation choices."],"supporting_citations":[{"why":"Supplies the Douglas string-equation construction of the FM-y corner with the polynomial $Q(y)=y^2+u_1$ for the Lee-Yang series.","marker":"[3]"},{"why":"Derives the resonance transformation from conformal selection rules using Jacobi polynomials; the master formula of Appendix A is reconciled with this construction.","marker":"[7]"},{"why":"Sets the universality filter that discards non-negative even $\\mu$-power correlators, the load-bearing projection the paper inherits.","marker":"[12]"},{"why":"Conjectures the $x\\leftrightarrow y$ swapped spectral curve and supplies the compact resonance formula (3.11) used for the spectral-curve and Frobenius $y$-side corners.","marker":"[15]"},{"why":"Proves the all-genus equivalence of the standard and swapped spectral curves; the paper's genus-zero frame-change realisation builds directly on this theorem.","marker":"[16]"},{"why":"Introduces the FM-x $A_{p-1}$ formulation without resonance and documents a four-point discrepancy that the present paper does not address.","marker":"[17]"},{"why":"Establishes KP integrability of the swapped spectral-curve tau-function, justifying the residue definition of the SC-x times.","marker":"[18]"},{"why":"Supplies the generalised Kontsevich weight with potential $\\Phi^p/p$ underlying the matrix-model origin of the $x\\leftrightarrow y$ swap.","marker":"[24]"}],"fun_headline_variants":["Four corners of (2,p) minimal gravity meet at genus zero","κ=-2 ties four dual descriptions of Lee-Yang minimal string","x↔y swap unifies four minimal gravity descriptions","Four roads to (2,p) minimal gravity: one tau-function"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The universality filter that discards correlators whose $\\mu$-power is a non-negative even integer is assumed rather than derived; it removes the fusion-forbidden amplitudes on the $y$-side corners, and without it the four-corner agreement would not hold.","fun_headline_variants_meta":{"raw":{"variants":["Four corners of (2,p) minimal gravity meet at genus zero","κ=-2 ties four dual descriptions of Lee-Yang minimal string","x↔y swap unifies four minimal gravity descriptions","Four roads to (2,p) minimal gravity: one tau-function"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000646,"raw_usage":{"total_tokens":3016,"prompt_tokens":1041,"completion_tokens":1975,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":657,"completion_tokens_details":{"reasoning_tokens":1902}},"tokens_in":657,"tokens_out":1975,"duration_ms":15287,"temperature":1.0,"reasoning_tokens":1902,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:28:01.342308+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the universal three-point ratio $R(1,1,1)$ for the (2,13) model in all four corners with $\\kappa=-2$; the claim predicts the CFT value $121/195$, and any corner that deviates from it disproves the equivalence. A second check is the fusion-forbidden triple $(1,1,3)$, which must vanish in all four corners; a non-zero value at (2,13) would mark the boundary of the dictionary.","supporting_citations":[{"cited_title":"Minimal Liouville Gravity correlation numbers from Douglas string equation","cited_arxiv_id":"1310.5659","evidence_quote":"Derives the resonance transformation from conformal selection rules using Jacobi polynomials; the master formula of Appendix A is reconciled with this construction."},{"cited_title":"$x-y$ swap for $(2,2p+1)$ minimal string","cited_arxiv_id":"2506.09222","evidence_quote":"Conjectures the $x\\leftrightarrow y$ swapped spectral curve and supplies the compact resonance formula (3.11) used for the spectral-curve and Frobenius $y$-side corners."},{"cited_title":"Dekinga, S","cited_arxiv_id":null,"evidence_quote":"Proves the all-genus equivalence of the standard and swapped spectral curves; the paper's genus-zero frame-change realisation builds directly on this theorem."}],"review_version":1}