{"id":"f046a06f-030b-4c8c-9c10-e4f199c19baa","arxiv_id":"2607.19462","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The Runkel-Watts string is a solvable family of two-dimensional string theories whose all-genus amplitudes are generated by SU(2) Yang-Mills TQFT data combined with VMS quantum volumes, and it has a dual matrix-integral description.","lead":"This paper introduces a new family of two-dimensional string theories, the Runkel-Watts string, and derives an all-genus formula for its amplitudes as a limit of the complex Liouville string. It also identifies a dual matrix-integral description and interprets the result as SU(2) Yang-Mills theory coupled to gravity.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Matrix-integral duality rests on the explicitly omitted topological-recursion reconstruction from (2.33) to (1.3); only low-order resolvents are checked.","rationale":"The reader's weakest_assumption already flags the skipped topological-recursion reconstruction, and the present pass agrees: this is the load-bearing gap in the central claim. The paper explicitly states it will not repeat the reconstruction, so the duality to a matrix integral is asserted rather than derived. Low-order checks are genuine evidence but do not cover the higher-order behavior of an infinite-sheeted, non-algebraic spectral curve with infinitely many branch points and nodal singularities. I therefore do not change the reader's CONDITIONAL verdict. I also noticed a concrete arithmetical inconsistency in the displayed evaluation of (2.54): Γ(3/2)√π (π√qq')² (-1/(2m)²) equals -π³ qq'/(8m²), not -1/(8π² qq' m²) as printed. If this is not a typographical error in the power of (π√qq') in (2.33), the derivation of the Feynman rules (2.48) itself would be affected. I did not make this the primary concern because the final amplitudes in §2.3 are numerically checked against direct integration, suggesting the displayed evaluation or the preceding edge factor contains a typo rather than a substantive error; nonetheless it should be corrected, and the concrete test above should be run with the corrected edge normalization.","tokens_in":31557,"tokens_out":29096,"duration_ms":263862,"concrete_test":"Implement the omitted reconstruction: for the RW spectral curve (2.78), compute from the topological recursion (2.82) the next unresolved differentials, e.g. ω_{2,0}, ω_{1,2}, and ω_{0,5}. Apply the dictionary (2.87) and compare term-by-term with the stable-graph sum (2.33) for RW_{2,0}, RW_{1,2}, and RW_{0,5}. Agreement to, say, 12 digits would close the gap; any mismatch would falsify the matrix-integral duality as stated. As a secondary check, verify the dilaton equation (2.94) at these orders.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim includes the statement that RW_{g,n} is generated by topological recursion from the spectral curve (1.3)/(2.73). The paper derives this curve in §2.5 by 'comparing (2.33) with the universal stable graph expansion' and then says: 'We will therefore not repeat the reconstruction in detail.' That reconstruction is precisely what would connect the intersection-theoretic Feynman rules to the matrix-integral duality. The checks in §2.5 cover only ω_{0,3}, ω_{1,1}, and ω_{0,4}; they do not exercise the infinite branch-point sum (2.82) in orders where the non-algebraic multi-sheet structure, the infinite branch points (2.74), or the nodal singularities (2.76) could contribute differently. The dilaton equation (2.94) and the exact-solvability claim both pass through this same identification. Thus the matrix-integral duality is a plausible but unproven assertion, not a derived consequence of the paper's arguments. The all-genus formula (2.48) may still be correct; the gap is specifically the identification of (2.33) with the topological recursion of (1.3).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a new family of two-dimensional string theories, the 'Runkel-Watts string', obtained by coupling Liouville CFT to the Runkel-Watts CFT. The central computational claim is an all-genus formula, Eq. (1.2)/(2.48), expressing string amplitudes as a sum over stable graphs with SU(2) Yang-Mills TQFT factors and Virasoro minimal string quantum volumes at the vertices. This formula is derived by taking a rational limit (2.31) of the complex Liouville string (CLS) intersection-theoretic Feynman rules (2.28). The paper also claims a dual description via topological recursion from a non-algebraic spectral curve (1.3)/(2.73), with a dictionary (2.87) between resolvents and Runkel-Watts amplitudes. Low-order results RW_{1,1} and RW_{0,4} are checked against direct numerical evaluation of the worldsheet CFT correlators, and several structural properties are established, including piecewise polynomiality, a triality symmetry, and a dilaton equation. In Section 3, the paper relates the Runkel-Watts string to the A-series minimal string and exhibits a coarse-grained factorization limit into a gauge theory factor times a VMS volume.","tokens_in":31910,"tokens_out":4869,"duration_ms":50853,"significance":"If the all-genus formula (2.48) is correct, this paper provides a genuinely new solvable string theory that sits between the complex Liouville string, the minimal strings, and the Virasoro minimal string, and whose amplitudes have elegant Feynman rules with SU(2) Yang-Mills data. The explicit low-order checks against numerical worldsheet CFT integration are a clear strength: they independently validate the limit (2.31) at the first nontrivial cases. The paper also correctly emphasizes that the VMS quantum volumes appear as universal building blocks, reinforcing the emerging structural picture in the field. However, the central claims rest on two load-bearing assumptions that are not fully proven: the interchange of the rational limit with the infinite sums/integrals in (2.28), and the identification of the resulting amplitudes with topological recursion from the spectral curve. The latter is explicitly left unproven ('We will therefore not repeat the reconstruction in detail'). These gaps do not invalidate the paper's evident plausibility, but they mean the all-genus and matrix-integral duality claims are not yet established at the level of proof.","major_comments":[{"comment":"The limit defining RW_{g,n} is taken on the CLS formula (2.28), which contains an infinite sum over stable graphs, an infinite sum over vertex colors, and an infinite edge expansion in d. Only the vertex factor is evaluated in the limit (2.32); the paper does not justify interchanging ε→0 with the graph sum, color sum, d-sum, and moduli-space integrals. The low-order checks in Sec. 2.3 cover only (g,n)=(1,1) and (0,4), which do not probe graphs with higher loop number or edge factors beyond the lowest d. Since the CLS formula is itself regularized and has pole-crossing subtleties, non-uniform convergence at rational b would alter (2.33) and hence (2.48). Please provide a justification of the interchange or add independent all-genus checks, such as RW_{2,0} or RW_{1,2}, evaluated directly from (2.33) and from the worldsheet CFT.","section":"Sec. 2.2, Eq. (2.31)"},{"comment":"The claimed matrix-integral duality rests on identifying (2.33) with topological recursion from the spectral curve (1.3)/(2.73). The paper explicitly declines to perform the reconstruction ('We will therefore not repeat the reconstruction in detail'). The checks (2.83)-(2.91) cover only ω_{0,3}, ω_{1,1}, and ω_{0,4}; these are the lowest orders and do not exercise the infinite branch-point sum (2.82) at orders where the non-algebraic branch points (2.74) or nodal singularities (2.76) could contribute. The dilaton equation (2.94) inherits the same gap. Consequently the duality is a conjecture supported by evidence, not a derived theorem. Please either present the reconstruction argument, even in summary, or verify the recursion at the next order (e.g., ω_{2,0} or ω_{1,2}) against an independent calculation.","section":"Sec. 2.5, Eq. (2.82)"}],"minor_comments":[{"comment":"The representation of the edge factor as a sum over ℓ_e with analytic continuation in s is delicate because the sum is not absolutely convergent at s=-1. The paper states an equivalence to the regularization in [14]; a few more details of why the ℓ-sum and the ψ-class expansion commute would improve readability.","section":"Sec. 2.2, Eq. (2.39)"},{"comment":"The chamber identity is reduced to 'finitely many cases' but the cases are not enumerated. The proof sketch would be easier to check if the list of cases (i)-(iii) were explicitly verified or tabulated.","section":"Appendix A, after Eq. (A.3)"},{"comment":"The inverse dictionary (2.88) is stated with a sign prescription; a brief comment on how the ±-signs are correlated in practice would prevent ambiguity, especially for non-symmetric integrands.","section":"Sec. 2.5, after Eq. (2.88)"},{"comment":"The limit from the A-series minimal string torus amplitude is verified only after 'fixing the overall normalization appropriately'; since this is a check of the relation, the normalization ambiguity should be spelled out more explicitly.","section":"Sec. 3.2, Eq. (3.14)"}],"recommendation":"major_revision","confidential_remarks":"This is a strong and interesting paper with a plausible central result. The main issue for me is that the matrix-integral duality is explicitly not derived, and the rational limit is not fully justified for all genus. If the authors can supply the reconstruction or a higher-order check (or a clear argument that the omitted reconstruction is a routine application of existing theorems), I would be happy to recommend acceptance. The paper is a good fit for a hep-th journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper introduces the Runkel-Watts string and solves it: an all-genus amplitude formula (2.48) with SU(2) Yang-Mills TQFT factors and VMS quantum volumes as vertices, obtained as a rational limit of the complex Liouville string. The formula is new and structurally striking, and the interpretation as 2d SU(2) Yang-Mills coupled to gravity is a nice payoff. The low-order checks — torus one-point and sphere four-point against direct numerical integration — are real evidence, and the agreement in the figures is convincing even without error bars.\n\nThe derivation is coherent as far as it goes. Starting from the CLS intersection formula, the ε→0 limit is handled graph-by-graph, colors rescale by q, and the Bernoulli/sign machinery trades color sums for momentum integrals. The piecewise polynomiality, absence of poles, and the dilaton equation are sensible consequences. The relation to the A-series minimal string via a Bézout-type limit is a welcome addition.\n\nThe main soft spot is exactly where the stress-test note points: the topological-recursion/matrix-integral duality. The paper says “We will therefore not repeat the reconstruction in detail” and instead compares (2.33) with the universal stable graph expansion to read off the spectral curve. That comparison is an assertion; the checks cover only ω0,3, ω1,1, and ω0,4. These low orders do not seriously exercise the infinite branch-point sum, the nodal singularities, or the multi-sheet structure. So the duality is plausible and low-order-verified, but not derived. A referee should ask for the reconstruction to be written out, or at least for a proof that the resolvent differentials from (1.3) satisfy the same recursive structure as (2.33).\n\nA second, smaller concern is the exchange of limits: the rational limit is taken inside an infinite sum over graphs and colors. The paper checks low orders numerically, but a uniformity argument is absent. If the CLS formula has unresolved subtleties near rational b, the RW formula inherits them. I would not call this fatal; the checks give reasonable confidence.\n\nThe paper is honest about inheriting CLS scaffolding, and the self-citation pattern is normal for a research program. This is a serious paper from a strong group. Anyone working on two-dimensional gravity, minimal strings, matrix models, or TQFT should read it. It deserves a serious referee, not a desk reject. My recommendation: send it to review, with instructions to focus on the topological-recursion identification and the limit interchange. If those survive, this is a major addition to the 2d string landscape.","headline":"New exactly solvable 2d string family with an elegant all-genus Feynman-rule formula and honest low-order checks; the matrix-integral duality is plausible but the key reconstruction step is skipped.","tokens_in":32371,"tokens_out":3139,"would_cite":true,"duration_ms":31660,"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 Runkel-Watts string has exact all-genus amplitudes: a stable-graph sum with SU(2) Yang-Mills factors at each vertex, reproduced by topological recursion from a spectral curve.","keywords":["Runkel-Watts string","complex Liouville string","minimal string theory","topological recursion","spectral curve","SU(2) Yang-Mills TQFT","Virasoro minimal string","matrix integral duality"],"falsifier":"Compute a higher-order Runkel-Watts amplitude directly from the worldsheet CFT, for example RW_{0,5} or RW_{1,2}, in the small-momentum regime where no contour deformation is needed, and compare with the prediction of (2.48). Any disagreement beyond numerical precision, or any dependence of the epsilon -> 0 result on the order of summation over colors versus edges, would falsify the central claim. A cheaper probe is to test the dilaton equation (2.94) at the next order (g,n) = (1,2), where only the contact graph contributes in the check.","tokens_in":31489,"feed_emoji":"🧵","tokens_out":7738,"duration_ms":74230,"temperature":0.7,"pith_summary":"This paper introduces a new family of two-dimensional string theories, the Runkel-Watts string, obtained by tensoring Liouville CFT with the generalized Runkel-Watts CFT and the bc-ghost system, and claims to solve them completely. Its central result is an all-genus formula for the string amplitudes: each amplitude is a finite sum over stable graphs, with a two-dimensional SU(2) Yang-Mills TQFT factor and a Virasoro-minimal-string quantum volume at each vertex, and a regularized momentum integral on each edge. The same amplitudes are claimed to be generated by topological recursion from an explicit non-algebraic spectral curve, giving a duality with a matrix integral. If the formula is correct, this family becomes a solvable laboratory bridging the complex Liouville string, the A-series minimal string, and the Virasoro minimal string, with amplitudes that are piecewise polynomial, pole-free, and satisfy a derivative-form dilaton equation.","feed_headline":"One formula solves the Runkel-Watts string at all genera","feed_subtitle":"The new 2d string family is exactly solvable: all amplitudes come from SU(2) Yang-Mills Feynman rules and a dual matrix integral.","key_machinery":"The load-bearing machinery is the regulated rational limit of the CLS intersection-theoretic stable-graph formula. Taking b to the rational value sqrt(q'/q) from the upper half-plane with an epsilon-scaled amplitude selects graphs whose vertex colors are multiples of q; after resumming colors via Bernoulli Fourier series, the formula becomes momentum-space Feynman rules in which the moduli-space content is packaged entirely into the VMS quantum volumes. The vertex factor is identified with SU(2) Yang-Mills theory through the character sum (2.45): A^{TQFT} equals a sum over SU(2) representations weighted by m^{-(2g-2+n)}, the same object that computes volumes of flat SU(2) moduli spaces. The","core_discovery":"The paper's central claim is that the Runkel-Watts string amplitudes RW_{g,n}(p) are given exactly by the stable-graph Feynman rules (2.48): summing over stable graphs with automorphism factors, each internal edge carries a regularized integral over a momentum k_e, and each vertex contributes the product of an SU(2) Yang-Mills TQFT partition function A^{TQFT}_{g_v,n_v}(sqrt(qq') p_v + (q+q')/2) with an analytically continued VMS quantum volume V^{sqrt(q'/q)}_{g_v,n_v}(ip_v). The paper derives this by taking a regulated rational limit b -> sqrt(q'/q) of the complex Liouville string amplitudes, in which only vertex colors divisible by q survive and the infinite color sums become Bernoulli poly","pith_inferences":["Editorial extension: the same limiting mechanism that turns the CLS stable-graph formula into SU(2) gauge times gravity Feynman rules would likely work for any rational degeneration of one Liouville factor; the resulting family would be labeled by the TQFT that replaces SU(2), making the gauge factor a universal fingerprint of the limit.","Editorial extension: at q=q'=1 the identification of momenta with SU(2) conjugacy classes and the flat-bundle volume interpretation suggests that the amplitudes could eventually be read as S-matrix-like asymptotic observables in a Lorentzian target space bounded by Runkel-Watts walls, a step the paper does not itself take.","Editorial extension: the derivative in the dilaton equation (2.94) indicates that naive insertion at the dilaton momentum vanishes in this theory; testing whether this non-commutation persists at higher genus would sharpen the dictionary between the worldsheet limit and the matrix-model resolvents.","Editorial extension: if the all-genus formula survives, the Runkel-Watts string would be the simplest example where a two-dimensional gauge theory, rather than a q-deformed gauge theory, is coupled to gravity through universal VMS vertices, suggesting a direct route from rational CFT limits to solvable gravity."],"forward_implications":["If the central formula holds, the Runkel-Watts string is exactly solvable at all genera: every amplitude RW_{g,n} is obtained from (2.48) by finitely many algebraic operations.","The same amplitudes are generated by topological recursion on the explicit spectral curve (2.73), so the theory is dual to a matrix integral whose eigenvalue density is the Chebyshev expression (2.77).","In the large-(q,q') limit the amplitudes factorize into an SU(2) Yang-Mills factor times a VMS quantum volume; the VMS volume is the envelope of a rapidly oscillating amplitude, giving a coarse-grained realization of timelike Liouville theory.","The Runkel-Watts string is a limit of the A-series minimal string, so the explicit Runkel-Watts amplitudes provide a solved corner of A-series minimal string theory and constraints on its general intersection-theoretic formula.","The amplitudes are piecewise polynomial with discontinuities but no poles, and satisfy a dilaton equation in which the dilaton insertion must be differentiated before evaluation, directly inherited from topological recursion."],"fun_headline_variants":["Exact amplitudes for the Runkel-Watts string at all genera","Runkel-Watts string solved: SU(2) gauge theory gives all amplitudes","New 2D string family: all-genus formula from Yang-Mills Feynman rules","Runkel-Watts string: one formula links all genera to matrix integral","2D string solved: amplitudes from Yang-Mills and a dual matrix model"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The all-genus formula rests on interchanging the rational limit epsilon -> 0 with infinite color sums, edge expansions, and moduli-space integrals in the complex-Liouville-string Feynman rules, and on the topological-recursion reconstruction of the Runkel-Watts spectral curve; the paper verifies the interchange only at low orders numerically and explicitly leaves the reconstruction underived.","fun_headline_variants_meta":{"raw":{"variants":["Exact amplitudes for the Runkel-Watts string at all genera","Runkel-Watts string solved: SU(2) gauge theory gives all amplitudes","New 2D string family: all-genus formula from Yang-Mills Feynman rules","Runkel-Watts string: one formula links all genera to matrix integral","2D string solved: amplitudes from Yang-Mills and a dual matrix model"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000146,"raw_usage":{"total_tokens":967,"prompt_tokens":640,"completion_tokens":327,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":384,"completion_tokens_details":{"reasoning_tokens":221}},"tokens_in":384,"tokens_out":327,"duration_ms":4178,"temperature":1.0,"reasoning_tokens":221,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T12:39:39.752447+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute a higher-order Runkel-Watts amplitude directly from the worldsheet CFT, for example RW_{0,5} or RW_{1,2}, in the small-momentum regime where no contour deformation is needed, and compare with the prediction of (2.48). Any disagreement beyond numerical precision, or any dependence of the epsilon -> 0 result on the order of summation over colors versus edges, would falsify the central claim. A cheaper probe is to test the dilaton equation (2.94) at the next order (g,n) = (1,2), where only the contact graph contributes in the check.","supporting_citations":[],"review_version":1}