{"id":"c37f4fe2-d366-473f-a616-9e0fcd209b34","arxiv_id":"2505.17129","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Multiple chordal SLE(0) systems of type (n,m) have traces given by the real locus of rational functions with n prescribed critical points and m poles, and their growth-point dynamics is the classical Calogero-Moser system.","lead":"The paper constructs deterministic zero-noise limits of multiple chordal SLE curves in a new configuration, where some curves connect boundary points pairwise and the rest run to a common point, and identifies the curves as the real contour lines of rational functions with prescribed critical points. It then shows the growth points move like particles in the classical Calogero-Moser system, linking probability theory, rational-function geometry, and integrable dynamics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The advertised exactness 'traces = real locus' is not proved: Theorem 1.5 contains only K_t ⊆ Γ(R), and the paper never establishes the reverse inclusion.","rationale":"The reader's weakest_assumption was Conjecture 3.1, the kappa-to-0 concentration of partition functions. That conjecture is unproved, but the paper explicitly says in Section 3.1 that the construction is self-consistent and does not depend on resolving the classical limit, so a failure of the conjecture would weaken the interpretation of 'SLE(0)' without invalidating the internal theorems. The trace converse, by contrast, is a hole in the central geometric claim as stated: the proof establishes containment, while the abstract and Section 1.2 assert equality. That makes the reverse-inclusion gap more load-bearing for the paper's main advertised result. A secondary issue noted by the reader is also real: Hamilton's equations for H=Σp_j^2/2 - 8Σ_{j<k}(x_j-x_k)^{-2} give x_j double dot = -16Σ_{k≠j}(x_j-x_k)^{-3}, not the -8 appearing in Theorem 4.2(ii); the coefficient in (1.13) or in the equations of motion should be adjusted. This is a genuine algebraic inconsistency, but it is independent of the trace characterization, so it does not displace the reverse-inclusion gap as the primary concern. The reader's CONDITIONAL verdict remains appropriate: the paper is a substantial, mostly direct computation extending the ABKM20 program, and the missing converse is a fixable but essential gap that should be stated honestly or proved.","tokens_in":24518,"tokens_out":12659,"duration_ms":116040,"concrete_test":"For the n=4,m=2 example of Figure 4.12 (x_1=i, x_2=e^{iπ/4}, x_3=e^{-iπ/4}, x_4=-i, ξ_{1,2}=0.32979±0.94405i), compute the full real locus Γ(R) by flowing along v_R=1/R' from all four critical points in both directions, and independently simulate the multiple Loewner flow (1.3)-(1.4) with ν_j=1 until collision, tracking the traces g_t^{-1}(x_j(t)). Compare the two sets in H. If every non-boundary arc of Γ(R) is covered by the traces, the missing converse is supported in this nontrivial case; if any arc of Γ(R) is absent from K_t, Theorem 1.5 is strictly one-sided and the abstract's 'correspond to' must be weakened to 'is contained in.' The same comparison can be repeated for a generic real x-configuration in H with a conjugate pair ξ solving the stationary relations (1.7).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 1.2 states that the real locus of R is 'exactly the traces' and the abstract repeats the correspondence, but the theorem that is actually proved (Theorem 1.5, Section 3.4) is one-sided: the hulls K_t generated by the multiple Loewner flow are a subset of Γ(R) up to collision times, and R composed with g_t^{-1} stays in C_R^{m,n}(x(t)). The integral of motion N_t(z) (Theorem 1.6) forces each growing trace to lie on a level curve Im R = 0, but it does not force every component of the real locus to be visited. The real locus of a real rational function can, in principle, contain arcs whose endpoints are poles or branches at critical points, and the Loewner dynamics selects particular flow directions from each starting point x_j. The reverse inclusion Γ(R) ∩ H ⊆ K_t (up to the same collision time) is never proved; the sentence before Theorem 1.5 simply asserts it. This matters directly for Theorem 1.9, where the classification of SLE(0) systems is declared equivalent to the rational functions and to critical points; without the converse, that equivalence overstates the proved result. The appended remark that the constructed systems are 'not known to encompass all possible' SLE(0) systems does not repair this gap, because the gap concerns the traces of the systems that are constructed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a deterministic multiple chordal Loewner evolution of type (n,m), called a multiple chordal SLE(0) system, defined through stationary relations that are the critical point equations of a master function. The main claims are (i) that the traces of these Loewner flows are exactly the real locus of a real rational function R with the driving points as simple critical points and the screening charges as simple poles; (ii) that this class of rational functions is classified by critical points of the rational KZ master function; and (iii) that, under a common capacity parametrization, the driving points evolve by the classical Calogero-Moser equations. The paper includes a field integral of motion N_t, examples in the disk uniformization, and a heuristic derivation of the construction as the kappa-to-0 limit of multiple chordal SLE(kappa).","tokens_in":24593,"tokens_out":11955,"duration_ms":95679,"significance":"If the trace identification were fully established, the paper would give a clean geometric description of deterministic multiple-SLE traces as contour lines of a single rational function, tying SLE(0) to the enumerative theory of real rational functions and to critical points of rational KZ equations. The Calogero-Moser dynamics is an elegant and nontrivial integrable-systems consequence, and the direct verification of the integral of motion and the explicit cancellation in the Calogero-Moser derivation are valuable. The examples illustrate a rich family of link patterns. However, the exactness of the trace correspondence is not proved, and the SLE(0) interpretation rests on a conjectural steepest-descent limit; these caveats substantially temper the paper's main significance.","major_comments":[{"comment":"The paper's central claim that traces are 'exactly the real locus' is not supported by the stated theorem. Theorem 1.5 asserts only that the hulls K_t are a subset of Γ(R) up to collision times and that R∘g_t^{-1} stays in C_R^{m,n}(x(t)); no reverse inclusion is stated or proved. The field integral of motion forces each growing trace to lie on a level set Im R = 0, but it does not force every component of the real locus to be traced: the real locus can contain arcs emanating from a critical point in the opposite flow direction or arcs ending at poles, and the Loewner dynamics selects particular flow directions from each starting point. The sentence immediately before Theorem 1.5 asserts the exact equality, and the abstract repeats it. This gap is load-bearing because Theorem 1.9 presents the classification of rational functions whose real locus forms a link pattern as a classification of SLE(0) systems, which is only justified if the reverse inclusion holds. The authors should either prove the converse or explicitly downgrade the correspondence to a containment statement and adjust the abstract and Theorem 1.9 accordingly.","section":"Section 1.2 and Theorem 1.5"},{"comment":"The proof of Theorem 1.5 is not written out. After stating the theorem, the text says only that the key ingredient is the integral of motion, then states Theorem 1.6 and Lemma 3.7; no argument is given showing how N_t implies K_t ⊆ Γ(R). Moreover, the use of the stationary relations at positive times in the proof of Theorem 4.2 (equations (4.1)-(4.2)) requires that the evolved configuration (x(t), ξ(t)) satisfy the stationary relations for all t, which is equivalent to the class invariance R∘g_t^{-1} ∈ C_R^{m,n}(x(t)) asserted in Theorem 1.5. Since that theorem is not proved, the Calogero-Moser derivation depends on an unproved preservation statement. The paper should provide a direct proof of the class invariance along the Loewner flow, or at least a separate lemma proving that the stationary relations are preserved.","section":"Sections 3.4 and 4.1"},{"comment":"The paper presents the constructed flows as 'multiple chordal SLE(0) systems' and, in the introduction and abstract, implies that they are the classical limit of multiple chordal SLE(kappa). However, the kappa-to-0 limit is explicitly conjectural: Conjecture 3.1 asserts that normalized partition functions concentrate on critical points of the master function, and the paper itself states that the construction is self-consistent and does not depend on resolving the classical limit. If the conjecture fails, the stationary-relations Loewner chains are a valid deterministic model but are not proven to be the SLE(0) limit of the random SLE(kappa) processes. This conditionality should be stated in the abstract or introduction, and the title's unqualified use of 'SLE(0)' should be hedged.","section":"Section 3.1 and Conjecture 3.1"}],"minor_comments":[{"comment":"The index in the stationary relations (1.1) is written as k=1,2,...,n, but the screening charges ξ_k are indexed by 1,...,m; it should read k=1,2,...,m, as in (1.7).","section":"Definition 1.1"},{"comment":"In equation (1.8), the factor g_t(x) in the denominator should be g_t(z); the same correction applies to the displayed formula in the proof of Lemma 3.7.","section":"Theorem 1.6"},{"comment":"The notation changes from b(W_t) to v(W_t) in the same derivation, and the term -3Ψ''_t(W_t) appears without a displayed factor; please check the formula against the cited Lawler reference and make the notation consistent.","section":"Section 2.1, equation (2.1)"},{"comment":"The summary proof uses the letter d for the number of poles while the theorem uses m, and the case range '0≤m≤n' should be consistent with the standing assumption m≤⌊n/2⌋; please unify the notation.","section":"Section 3.3, proof of Theorem 1.4"},{"comment":"The manuscript refers to Figures 1-15, but the actual figures are not included in the text; since the examples and the claimed link patterns are verified visually in these figures, their absence makes those sections difficult to assess.","section":"Section 3.7 and 3.8"},{"comment":"The displayed computation of ∑ x_j U_j contains corrupted notation (a stray '+' and a missing expression); the final result appears correct, but the line should be cleaned up.","section":"Proof of Theorem 3.6"},{"comment":"The entry [MZ24a] is listed twice with different titles, and [JZ25a] is a self-reference to the present paper; please renumber and deduplicate the bibliography.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The main gap is the missing converse in the trace theorem and the unproved preservation of the stationary-relations class under the flow. If the author can prove the reverse inclusion or explicitly weaken the main claim, the paper's core computations would remain valuable. The paper would also benefit from a clear separation between the conjectural κ→0 limit and the definition of the deterministic Loewner chains. Given the paper's reliance on companion preprints, the editor may wish to verify that [ABKM20] and the author's other self-cited works are in a suitable form for reference."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a real extension of the ABKM20 multiple SLE(0) construction from n = 2m to general type (n,m) with m ≤ floor(n/2). What is genuinely new: the field integral of motion N_t(z) for the chordal Loewner flow, the containment theorem for hulls in the real locus of a rational function, and the derivation of Calogero-Moser dynamics for the growth points under common capacity parametrization. The bulk of the proofs are direct, explicit computations, and the stated classification theorems import cleanly from EG02 and SV03. There are no fitted parameters and no self-citation-driven circularity.\n\nThe soft spots are real but not fatal. First, the advertised correspondence is stronger than what is proved. Theorem 1.5 gives only K_t ⊆ Γ(R); the reverse inclusion is asserted but not established. The stress-test note is right: the integral of motion forces each growing trace to lie on a level curve, but does not force every component of the real locus to be visited. So the abstract's \"traces correspond to the real locus\" overstates the proved result. Second, Conjecture 3.1 — the steepest-descent limit identifying the stationary relations with the kappa-to-0 limit of SLE(kappa) — is unproved and is the premise that fixes these as the \"physical\" SLE(0) dynamics. The author is honest about this and says the construction is self-consistent, but the significance claim hangs on that conjecture. Third, the Calogero-Moser Hamiltonian in (1.13) has coupling 8, while the derived equations of motion in Theorem 4.2(ii) require coupling 4 under the Hamiltonian equations as stated. That is a concrete inconsistency, worth fixing. Minor issues: the abstract mentions radial SLE(0) without a corresponding theorem, the Poisson bracket computation in the proof of (1.14) is not shown, and the numerical examples in Section 3.8 lack precision or error estimates.\n\nOverall, the core construction is coherent and the computations check out at the level of explicit algebra. The paper would benefit from a careful revision that (a) proves or clearly delimits the trace inclusion, (b) fixes the Hamiltonian coupling constant, and (c) states Conjecture 3.1 as a genuine open problem rather than a background assumption. Who is this for? Researchers working on deterministic limits of multiple SLE and on Loewner dynamics with integrable structure. It deserves a serious referee: the work is substantial and mostly correct, but not ready in its current form.","headline":"Genuine extension of the ABKM20 SLE(0) program to general (n,m), with real content in the integral-of-motion and Calogero-Moser results, but the advertised trace correspondence is only proved one-way.","tokens_in":25363,"tokens_out":1768,"would_cite":true,"duration_ms":17213,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60J67","37J35","30C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"The traces of multiple chordal SLE(0) systems are the real locus of a single rational function fixed by the starting points, and their equal-capacity dynamics are the classical Calogero-Moser system.","keywords":["multiple SLE(0)","Loewner evolution","rational functions with prescribed critical points","Calogero-Moser system","Knizhnik-Zamolodchikov master function","stationary relations","integral of motion","classical limit of SLE"],"falsifier":"Numerically integrate a small case at the predicted boundary: pick generic real starting points $x$ for, say, $(n,m) = (3,1)$, solve the stationary relations for the screening charge $\\xi$, plot the real locus of $R$ with $R'(z) = c\\prod (z-x_j)/((z-\\xi)^2 (z+1)^{n-2m+1})$, and run the Loewner flow with $\\nu_j = 1$. If the traced hull fails to cover all of $\\Gamma(R)$, or leaves it before a collision, the identification in Theorem 1.5 is only a containment. To test the $\\kappa \\to 0$ premise itself, simulate the driving functions of the chordal SLE($\\kappa$) multiple system at very small $\\kappa$ for fixed $x$ and check whether the empirical concentration points of the partition functions approach the master-function critical points predicted by the stationary relations.","tokens_in":24090,"feed_emoji":"📐","tokens_out":21099,"duration_ms":157787,"temperature":0.7,"pith_summary":"Random Schramm-Loewner curves SLE($\\kappa$) are expected to freeze into fixed deterministic curves as the parameter $\\kappa$ goes to 0, and the balanced case — $2n$ boundary points paired into $n$ curves — was already understood. This paper builds the general type-$(n,m)$ theory, in which $n$ starting points on the real line produce $m$ non-crossing curves pairing some of the points while the remaining $n-2m$ curves run off to the marked point at infinity. The main claim is geometric: the whole collection of traces is the real locus of a single real rational function — the set of points where the function takes real values — whose $n$ simple critical points are the starting points and whose $m$ simple poles are the screening charges, with a pole of order $n-2m+1$ at infinity. The second claim is dynamical: under a common capacity parametrization the starting points move like particles of the classical Calogero-Moser system with inverse-square repulsion. If both hold, the deterministic limit of multi-curve SLE is captured exactly by one rational function and one integrable particle system.","feed_headline":"Deterministic SLE(0) traces reduce to one rational function's real locus","feed_subtitle":"The κ→0 limit of multi-curve SLE obeys Calogero-Moser dynamics, with traces on one rational function's real locus.","key_machinery":"The argument is carried by a preserved differential, the field integral of motion $N_t(z)$: with the marked point at infinity, $N_t(z) = g'_t(z)\\big/\\big(\\prod_{k=1}^n (g_t(z)-x_k(t)) \\prod_{j=1}^m (g_t(z)-\\xi_j(t))^2\\big)$, and with a finite marked point $u$ an extra factor $(g_t(z)-u)^{2m-n-2}$ appears. Along the multiple Loewner flow the logarithmic derivative of $N_t(z)$ vanishes identically, so the growing conformal map $g_t$ continues to conjugate the same meromorphic differential; from this the paper derives that the traced hulls lie on the real locus of a rational function $R$ and that the transported function $R \\circ g_t^{-1}$ stays in $C_R^{m,n}(x(t))$, with the moving critical points $x(t)$. The complementary machinery is algebraic: the stationary relations (equivalently, criticality of the master function), the quadratic null-vector equations $\\frac{1}{2} U_j^2 + \\sum_{k \\neq j} \\frac{2}{x_k-x_j} U_k - \\sum_{k \\neq j} \\frac{6}{(x_j-x_k)^2} = 0$, and the Calogero-Moser Lax pair $(L,M)$, which turns the null-vector equations into $L^2 \\mathbf{1} = 0$ and shows that $H = \\sum_j H_j$ is the classical Calogero-Moser Hamiltonian.","core_discovery":"The central correspondence (Theorem 1.9) identifies three equivalent descriptions of a type-$(n,m)$ chordal SLE(0) system: a multiple Loewner flow driven by the stationary relations between growth points $x = (x_1,\\dots,x_n)$ on the real line and conjugation-symmetric screening charges $\\xi = (\\xi_1,\\dots,\\xi_m)$, which are exactly the critical point equations of the rational Knizhnik-Zamolodchikov master function $\\Phi(x,\\xi)$; a real rational function $R$ with simple critical points at $x$, simple poles at $\\xi$, and a pole of order $n-2m+1$ at infinity, whose real locus $\\Gamma(R)$ carries the $(n,m)$ link pattern; and the corresponding critical point of the master function. The trace theorem (Theorem 1.5) proves that the hulls $K_t$ generated by the Loewner flow stay inside $\\Gamma(R)$ up to collision times, with $R \\circ g_t^{-1}$ remaining in the class $C_R^{m,n}(x(t))$; the paper states the intended conclusion as the real locus being exactly the traces. Dynamically (Theorems 1.10 and 4.2), with equal capacity parametrization $\\nu_j = 1$ the growth points satisfy $\\ddot{x}_j = -\\sum_{k \\neq j} \\frac{8}{(x_j-x_k)^3}$, the classical Calogero-Moser equations with coupling 8; the null-vector Hamiltonians $H_j$ sum to the Calogero-Moser Hamiltonian $H = \\sum p_j^2/2 - \\sum_{j<k} \\frac{8}{(x_j-x_k)^2}$, they Poisson-commute along the invariant manifolds $N_c$, and the Lax pair $(L,M)$ encodes the constraints as $L^2 \\mathbf{1} = 0$.","pith_inferences":["A low-cost probe of the whole theory is a very small-$\\kappa$ simulation of a single link pattern: the predicted screening charge $\\xi$ is computable from the stationary relations, so the empirical concentration point of the SLE($\\kappa$) partition function either matches it or refutes Conjecture 3.1 without needing to resolve the traces.","Calogero-Moser integrability implies higher conserved quantities, such as traces of powers of the Lax matrix $L$, beyond those written down; whether these quantities are preserved along the $\\nu_j = 1$ Loewner flow is not addressed and can be checked directly.","If the trace identification holds in full, the SLE(0) curve ensemble is compressed into finitely many real parameters — the starting points and the additive constant of $R$ — which suggests a low-dimensional, welding-style reconstruction of deterministic multiple-SLE traces and a concrete moduli count for over-screening families."],"forward_implications":["Every admissible $(n,m)$ link pattern is realized geometrically: in the under-screening regime $n+1-m > m$ by a unique rational function (up to a real constant) for generic $x$, and in the over-screening regime by a continuous family, so the classification of chordal SLE(0) systems is the classification of these rational functions.","At the threshold $n+1-m = m$ no such systems exist for generic data, so the SLE(0) construction has a sharp boundary in parameter space.","With equal capacity parametrization the growth points satisfy the classical Calogero-Moser equations $\\ddot{x}_j = -\\sum_{k\\neq j} \\frac{8}{(x_j-x_k)^3}$ with total energy zero, so the Loewner driving dynamics are integrable, with the null-vector Hamiltonians commuting along the invariant manifolds $N_c$ and an explicit Lax pair.","The equivalence of Theorem 1.9 transfers the enumeration of chordal SLE(0) systems to the known count of critical points of the rational KZ master function, and the integral of motion arises as the $\\kappa \\to 0$ limit of an SLE($\\kappa$) martingale built from vertex operators and screening charges."],"supporting_citations":[{"why":"The $n = 2m$ theory of multiple chordal SLE(0) with pole dynamics and the integral of motion; this paper extends that construction to all $(n,m)$ with $m \\le \\lfloor n/2 \\rfloor$.","marker":"[ABKM20]"},{"why":"Supplies the existence and enumeration of rational functions with prescribed real critical points that realize link patterns, used in the under-screening and over-screening classification.","marker":"[EG02]"},{"why":"Counts the critical points of the rational KZ master function, giving the enumeration and the threshold behavior in the classification.","marker":"[SV03]"},{"why":"Provides the vertex-operator and screening-charge conformal field theory from which the martingale observable — and its $\\kappa \\to 0$ limit, the integral of motion — is constructed.","marker":"[KM13]"},{"why":"Supplies the boundary CFT formalism used in Section 3.5 to derive the martingale observable for the chordal SLE($\\kappa$) system.","marker":"[KM21]"},{"why":"Gives the Lax-pair representation of the Calogero-Moser system used in Theorem 4.3 for the null-vector Hamiltonians.","marker":"[Mos75]"},{"why":"The precedent for the SLE/integrable-system correspondence — radial SLE($\\kappa$) with the quantum Calogero-Sutherland system — that the paper's classical Calogero-Moser result parallels.","marker":"[DC07]"}],"fun_headline_variants":["SLE(0) traces become rational function real loci, Calogero-Moser ensues","Multiple SLE(0) curves boil down to Calogero-Moser dynamics","Rational functions encode trace geometry in SLE(0) systems","SLE(0) link patterns live on rational function real loci","kappa=0 SLE traces follow Calogero-Moser equations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is Conjecture 3.1, that as $\\kappa \\to 0$ the normalized multiple SLE($\\kappa$) partition functions concentrate by steepest descent on critical points of the master function — without it the stationary-relation flows are only a self-consistent model, not the genuine SLE(0) limit — and, for the trace identification, the unproved reverse containment, since Theorem 1.5 establishes only that the hulls lie inside the real locus and not that the real locus is fully traced.","fun_headline_variants_meta":{"raw":{"variants":["SLE(0) traces become rational function real loci, Calogero-Moser ensues","Multiple SLE(0) curves boil down to Calogero-Moser dynamics","Rational functions encode trace geometry in SLE(0) systems","SLE(0) link patterns live on rational function real loci","kappa=0 SLE traces follow Calogero-Moser equations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000557,"raw_usage":{"total_tokens":2746,"prompt_tokens":1137,"completion_tokens":1609,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":753,"completion_tokens_details":{"reasoning_tokens":1513}},"tokens_in":753,"tokens_out":1609,"duration_ms":10740,"temperature":1.0,"reasoning_tokens":1513,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T15:08:53.274102+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically integrate a small case at the predicted boundary: pick generic real starting points $x$ for, say, $(n,m) = (3,1)$, solve the stationary relations for the screening charge $\\xi$, plot the real locus of $R$ with $R'(z) = c\\prod (z-x_j)/((z-\\xi)^2 (z+1)^{n-2m+1})$, and run the Loewner flow with $\\nu_j = 1$. If the traced hull fails to cover all of $\\Gamma(R)$, or leaves it before a collision, the identification in Theorem 1.5 is only a containment. To test the $\\kappa \\to 0$ premise itself, simulate the driving functions of the chordal SLE($\\kappa$) multiple system at very small $\\kappa$ for fixed $x$ and check whether the empirical concentration points of the partition functions approach the master-function critical points predicted by the stationary relations.","supporting_citations":[],"review_version":1}