{"id":"bd03d32e-c3f1-46e0-951f-cc5e4b003660","arxiv_id":"2504.14595","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Multiple SLE_kappa((kappa-6)/2,(kappa-6)/2) is shown to be unique and its hitting points, after collapsing starting points, form beta-Jacobi ensembles with beta=8/kappa, yielding a new scaling limit for critical Ising interfaces.","lead":"The paper constructs two versions of a multi-curve random process called multiple SLE_kappa((kappa-6)/2,(kappa-6)/2), proves they have the same law, and shows their hitting points on a boundary arc match beta-Jacobi ensembles with beta=8/kappa. It then uses this process to compute the probability that all critical Ising interfaces end on the free boundary arc and to identify their joint scaling limit.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Corollary 1.8's beta-Jacobi parameters contradict a direct change of variables from the Theorem 1.5/1.6 endpoint densities; for N=2, kappa=3 the claimed exponents 1/2,3/2 differ from the derived -1/3,1.","rationale":"The reader's conditional verdict focuses on the sketched uniqueness proof, and I agree that Lemma 3.4 and the modified Zhan argument are under-supported. But the more decisive, checkable problem is the stated Jacobi parameters. This is not a matter of missing rigor: a direct one-dimensional change of variables from the paper's own densities disproves Corollary 1.8 as stated for the minimal case N=2. The introduction even says the convergence to Jacobi is clear from the densities, so the parameters should be read off from exactly the computation above; they are not. The beta=8/kappa exponent survives, so the qualitative connection may be salvageable with corrected exponents, and the two flow-line constructions might still be valid. Hence I do not move the verdict to reject; the appropriate outcome remains conditional, now with the additional required correction of Corollary 1.8 and a strengthening of the uniqueness proof.","tokens_in":32800,"tokens_out":33713,"duration_ms":296036,"concrete_test":"Set N=2 and kappa=3. Take the density from Theorem 1.5: f(z) proportional to z^{-8/3}(z-1) on (1,infinity), apply y=1/z, and normalize to obtain g(y) proportional to y^{-1/3}(1-y) on (0,1). Compare g with the density claimed in Corollary 1.8, which is proportional to y^{1/2}(1-y)^{3/2}; for instance, E[y] under g is 1/4, while under the claimed density it is 3/8. Repeat the same change of variables with the Theorem 1.6 density for the even-offset hitting point. If the two transformed densities match the claimed Jacobi parameters, the objection is void; otherwise Corollary 1.8 must be corrected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Corollary 1.8 is internally inconsistent with the constructions it cites. For N=2, M=1, Theorem 1.5 assigns the odd-offset hitting point z the density C 1{z>1} z^{-8/kappa}(z-1)^{(6-kappa)/kappa}. With y=psi(z)=1/z, the density in y is C' y^{(2-kappa)/kappa}(1-y)^{(6-kappa)/kappa}, which by the definition in Section 2.3 is Jacobi(1;8/kappa,(2-kappa)/kappa,(6-kappa)/kappa). Corollary 1.8 instead claims Jacobi(1;8/kappa,1/2,3/2). The equations (2-kappa)/kappa=1/2 and (6-kappa)/kappa=3/2 have no common solution (kappa=4/3 versus kappa=12/5); at the Ising value kappa=3 the derived exponents are -1/3 and 1. The same computation from Theorem 1.6 gives y^{(6-kappa)/kappa}(1-y)^{(2-kappa)/kappa} for the even-offset hitting point, again contradicting the stated (3/2,1/2). Thus the parameter identification advertised in the abstract is wrong as stated, although beta=8/kappa itself is unchanged; the correct boundary exponents are the Selberg/Coulomb exponents visible in (3.6) and (3.15). Because Corollary 1.8 is stated as a theorem and used to advertise the beta-Jacobi connection, this is a load-bearing error independent of the uniqueness proof.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a family of multiple chordal SLEs with two force points of weight (κ-6)/2, called multiple N-SLE_κ((κ-6)/2,(κ-6)/2), and gives two GFF-flow-line constructions with explicitly sampled hitting points (Theorems 1.1, 1.2, 1.5, 1.6). It claims uniqueness of this multiple SLE from its conditional law (Theorems 1.3 and 1.7), derives the joint law of the hitting points on the target arc (Corollaries 1.4 and 1.8), and claims that after shrinking all starting points to one point the hitting points converge to β-Jacobi ensembles with β=8/κ. The paper then applies these objects to the critical Ising model: Theorem 1.9 gives the asymptotic probability of the event that all N discrete interfaces end on the free arc, and the conditional convergence of the interface tuple to the multiple SLE with κ=3.","tokens_in":33166,"tokens_out":13431,"duration_ms":112074,"significance":"If the main claims are correct, this paper would provide an explicit, conformally invariant endpoint distribution for a nontrivial family of multiple SLEs and a direct SLE-to-β-Jacobi connection, in addition to a rigorous conditional scaling limit for multiple critical Ising interfaces with mixed boundary conditions. The cascade computations in Section 3, especially the partition-function identity (3.19), are explicit and checkable, and the constructions via GFF flow lines with densities (1.3)–(1.6) are concrete enough to test directly. However, the advertised Jacobi corollary is false as stated, and the uniqueness arguments that support the endpoint laws are incomplete. These issues are load-bearing for the paper's central claims, so the significance is conditional on substantial revision.","major_comments":[{"comment":"A direct change of variables contradicts the stated Jacobi parameters. For N=2, M=1, the density (1.5) with x=0, u=1 is C 1_{1<z_1} z_1^{-8/κ}(z_1-1)^{(6-κ)/κ}. Setting y=ψ(z_1)=1/z_1 gives density C' y^{(2-κ)/κ}(1-y)^{(6-κ)/κ}, which by the definition in Section 2.3 is Jacobi(1;8/κ,(2-κ)/κ,(6-κ)/κ), not Jacobi(1;8/κ,1/2,3/2) as claimed. The equations (2-κ)/κ=1/2 and (6-κ)/κ=3/2 have no common solution; at the Ising value κ=3 the derived exponents are -1/3 and 1. In general the odd-endpoint density in the y=1/z coordinates is proportional to ∏_j y_j^{(4N-8M+2-κ)/κ}(1-y_j)^{(6-κ)/κ}∏_{i<j}|y_i-y_j|^{8/κ}, and the even-endpoint density is proportional to ∏_j y_j^{(4N-8(N-M)+6-κ)/κ}(1-y_j)^{(2-κ)/κ}∏_{i<j}|y_i-y_j|^{8/κ}. Corollary 1.8 is therefore not a minor misprint but a wrong parameter identification; the advertised SLE–β-Jacobi connection must be corrected (the relation β=8/κ itself is unchanged).","section":"Section 4, Corollary 1.8"},{"comment":"The uniqueness theorem is load-bearing for Corollary 1.4 and Theorem 1.7, but as written the proof is only an outline. After Lemma 3.3 the text states: \"The proof is almost the same as the proof of [Zha24, Theorem 4.2] and we only give the outline below.\" The key positivity estimate (3.20) is not proved in the manuscript: the three-step construction of the sets A_j yields (3.22) only if Lemma 3.4 supplies the required positive transition probability, but Lemma 3.4 is dispatched by saying \"This is same as the proof of [Yu23, Lemma 3.1]\", without checking that the cited hypotheses apply to SLE_κ((κ-6)/2,(κ-6)/2) in the required tubes and without proving the claimed absolute continuity of the corresponding flow lines on subdomains. Since the identification of the two constructions and the hitting-point formulas both rely on uniqueness, this gap is central rather than cosmetic.","section":"Section 3, Proof of Theorem 1.3"},{"comment":"The same-start uniqueness proof has two serious gaps affecting Corollary 1.8. First, the passage from (4.8) to (4.9) invokes \"monotone convergence\" for the integrals of ∂_{v_2} log U(g_s(x_-), W_s, g_s(u)); no monotonicity or sign property of this integrand is shown, and the stopping times σ_{ϵ0} depend on ϵ0, so the stated almost-sure limit is not justified. Second, the proof of Theorems 1.5 and 1.6 says \"the same proof still works\" when x_1=...=x_N, but the cascade lemmas in Section 3 then involve force points that are no longer distinct; the manuscript does not prove that the flow-line hitting descriptions from Section 2.2, or the defining conditional-law property, survive this degeneration. Both issues are load-bearing because Corollary 1.8 rests on Theorem 1.7.","section":"Section 4, Proof of Theorem 1.7"},{"comment":"There is an unflagged inconsistency in the announced value of κ for the Ising application. The paragraph before Theorem 1.9 says the conditional interface tuple converges to the multiple N-SLE with \"κ = 8/3\", while Theorem 1.9 itself, the abstract, and all of Section 5 (for example Lemma 5.4 with h=1/2 and κ=3) use κ=3. Since the choice of κ determines which SLE is the claimed scaling limit, this contradiction must be resolved.","section":"Section 1.2 and Theorem 1.9"}],"minor_comments":[{"comment":"The statement says \"ϕ(x_{2N+2})=∞\", which should presumably be ϕ(x_{N+2})=∞; the subscript appears to be a typo.","section":"Theorem 1.9 statement"},{"comment":"In the displayed inclusions, the term ∂B(x^δ_{N+1},ϵ) appears twice where the second occurrence should involve x^δ_{N+2}.","section":"Lemma 5.3"},{"comment":"Figure 1.4 and Figure 1.5 are captioned as illustrations of Theorem 1.1, but they illustrate the same-start constructions of Theorems 1.5 and 1.6 respectively.","section":"Figure captions"},{"comment":"In the bullet point for Q_N, the sentence \"η_{m-1} is on the left to η_m for 2≤ℓ≤N\" uses the wrong index ℓ; it should be \"for 2≤m≤N\".","section":"Section 2.2, properties of Q_N"},{"comment":"The reference [Yu23] has an incomplete title: \"Time-reversal of multiple-force-point chordal .\" ends with a dangling period; the missing word should be supplied.","section":"References"},{"comment":"There is a typo in \"Recall the definitio of partition function\"; it should be \"definition\".","section":"Section 1.2"}],"recommendation":"major_revision","confidential_remarks":"The parameter mismatch in Corollary 1.8 is directly checkable from the densities (1.5)-(1.6) and should be corrected before the paper can be considered further. I would also ask the editor to insist that the authors either supply a complete proof of Theorem 1.3 and Theorem 1.7 or explicitly downgrade these results to conditional statements, because the advertised endpoint and Jacobi claims depend on them."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nRead this one carefully before citing the β-Jacobi part. The stress-test note is right: Corollary 1.8 does not follow from the paper's own densities. For N=2, M=1, Theorem 1.5 gives the odd hitting point z density z^{-8/κ}(z-1)^{(6-κ)/κ} up to normalisation. Under y=1/z that becomes y^{(2-κ)/κ}(1-y)^{(6-κ)/κ}, i.e. Jacobi(1;8/κ,(2-κ)/κ,(6-κ)/κ) in the paper's own definition. Corollary 1.8 claims (1/2,3/2). No value of κ makes those equal; at κ=3 the derived exponents are -1/3 and 1. Same with the even-index density from Theorem 1.6. So the advertised direct connection to the β-Jacobi ensemble is wrong as stated. β=8/κ survives, but the boundary exponents need to be corrected.\n\nThat said, the paper has real content. The two GFF flow-line constructions for multiple N-SLE_κ((κ-6)/2,(κ-6)/2), and the same-start-point variants, are genuinely new. The cascade computation in Section 3 is coherent, and the induction in Lemma 3.3 that P_N=Q_N is plausible; I did not find a gap there. The explicit hitting-point densities (1.3)-(1.6) are useful even if the Jacobi labeling is wrong.\n\nThe soft spots are concentrated where the reader found them. Theorem 1.3's uniqueness is an outline that modifies Zhan's argument at the point where [BPW21, Lemma 3.6] fails, and the key absolute-continuity Lemma 3.4 is imported from [Yu23]. That is acceptable for a preprint, but the uniqueness theorem is load-bearing, so a referee should ask for a complete proof. Theorem 1.9 also has an unflagged κ=8/3 in the introduction against κ=3 in the statement and proof; the latter is the right Ising value, so it is likely a typo. Lemma 5.2 and Lemma 5.4 are asserted with sketches and references; they need more detail before the Ising application can be called solid.\n\nBottom line: this deserves a serious referee, not a desk rejection. The constructions are worth publishing, and the main error is contained in the Jacobi corollary and the advertised connection. I would send it to review with a request to fix the parameter identification and to write out the uniqueness argument. As it stands I would not cite Corollary 1.8, and I would not bring the paper to reading group until the Jacobi exponents are corrected.","headline":"A serious SLE/Ising paper whose advertised β-Jacobi identification does not survive a change of variables; the constructions are real, but Corollary 1.8 needs a major correction.","tokens_in":33700,"tokens_out":10119,"would_cite":false,"duration_ms":82917,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60J67"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that multiple $N$-SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$ is unique, builds it twice from GFF flow lines, identifies its coalesced-start hitting points with $\\beta$-Jacobi ensembles at $\\beta=8/\\kappa$, and shows…","keywords":["Ising model","Schramm-Loewner evolution","Gaussian free field flow lines","β-Jacobi ensemble","multiple SLE","conditional law uniqueness","hitting points","Selberg integral"],"falsifier":"For $N=2$, run the two couplings $P_2$ and $Q_2$ directly from the densities (1.3) and (1.4), simulate the GFF flow lines, and compare the distribution of the hitting point of the second curve: any difference would disprove the claimed equality of laws (Theorem 1.3). Alternatively, check positivity (3.20) for two tubes brought arbitrarily close to tangency; a zero transition probability would localize the failure in Lemma 3.4.","tokens_in":32549,"feed_emoji":"📐","tokens_out":15538,"duration_ms":117385,"temperature":0.7,"pith_summary":"At the center of the paper is a family of $N$ non-crossing random curves, called multiple $N$-SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$, whose defining property is that each curve, conditioned on the others, is a chordal SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$ in the remaining domain. The paper's main claim is that this conditional-law property actually fixes the law uniquely, so any two constructions that satisfy it must agree. Two such constructions are given using flow lines of the Gaussian free field with random marked points, and the paper proves they coincide. Exploiting that coincidence, it reads off the joint law of the points where the curves hit the target boundary arc, and after all $N$ starting points are collapsed into one point, those hitting-point laws are exactly $\\beta$-Jacobi ensembles with $\\beta=8/\\kappa$, establishing a direct SLE-to-random-matrix connection. In the critical Ising model with alternating and free boundary conditions, the same machinery produces an explicit conformal formula for the probability that all $N$ interfaces reach the free arc, and shows that, conditioned on that event, the interfaces converge to the $\\kappa=3$ member of the family.","feed_headline":"Two flow-line constructions give one SLE family with Jacobi endpoints","feed_subtitle":"Uniqueness by conditional law yields explicit Ising interface probabilities and a β=8/κ random-matrix link.","key_machinery":"The carrying mechanism is the imaginary-geometry flow-line construction of the Gaussian free field. The random input is a set of auxiliary boundary points $z_j$ or $w_j$ drawn from the explicit densities (1.3)/(1.4) and their same-start variants (1.5)/(1.6); the exponents in those densities, $-4/\\kappa$, $(6-\\kappa)/\\kappa$, $(2-\\kappa)/\\kappa$ and $8/\\kappa$, are exactly what makes the integrals Selberg-type and ultimately produces the Jacobi-ensemble parameters. The GFF is assigned stepwise boundary data $\\lambda-2N\\lambda$, $\\lambda(1-2N+2(j-1))$, and so on, at the starting points $x_j$ and at the sampled points; the $N$ curves are the flow lines issued from $x_j$ or from the common point $x$ with angles $2\\lambda(j-1)/\\chi$. The cascade relations in Lemmas 3.1 and 3.2 express the law of the last curve as a fixed SLE weighted by a martingale involving the partition functions $Z_N$ and $W_N$, and they identify the conditional law of the remaining $N-1$ curves recursively; the partition-function identity (3.19) then forces the two couplings $P_N$ and $Q_N$ to coincide. Uniqueness is proved by a Markov chain that resamples one curve at a time from its defining conditional law, with the hard positivity estimate (3.20) supplied by absolute continuity of the flow lines on subdomains (Lemma 3.4).","core_discovery":"Fix $N\\ge 1$ and $\\kappa\\in(0,4)$. The paper calls a probability measure on $N$ disjoint simple curves, running from prescribed boundary points to a common boundary arc, a multiple $N$-SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$ if each curve's conditional law given the others is chordal SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$ in the domain left by the other curves, with the neighbouring endpoints as force points. The central discovery is that this defining property is enough: the measure is unique when it exists (Theorem 1.3), and likewise for the variant where all curves start at the same point (Theorem 1.7). The paper then constructs the measure explicitly. It samples $M$ or $N-M$ auxiliary points on the boundary arc from the Selberg-type densities (1.3) or (1.4), gives the Gaussian free field piecewise-constant boundary data that jumps by multiples of $\\lambda=\\pi/\\sqrt{\\kappa}$ at the starting points and at the sampled points, and takes the $N$ flow lines with angles $2\\lambda(j-1)/\\chi$, where $\\chi=2/\\sqrt{\\kappa}-\\sqrt{\\kappa}/2$. These two constructions have the same law (Lemma 3.3), so the hitting points of the odd-indexed curves are distributed by (1.3) and those of the even-indexed curves by (1.4) (Corollary 1.4). When all starting points collapse to one point, inversion turns these laws into the $\\operatorname{Jacobi}(M;8/\\kappa,\\ldots)$ and $\\operatorname{Jacobi}(N-M;8/\\kappa,\\ldots)$ ensembles (Corollary 1.8).","pith_inferences":["Inference: if the uniqueness argument is sound, the same Markov-chain/positivity strategy should characterize other chordal multiple SLE$_\\kappa(\\rho)$ families with fixed extreme force points, wherever the relevant flow-line restriction is absolutely continuous.","Inference: the exact appearance of $\\beta$-Jacobi ensembles suggests that GFF flow-line configurations with marked boundary points form a natural ambient space for random-matrix spectra; a testable extension would be to couple the $z_j/w_j$ points to the eigenvalues of the tridiagonal Jacobi matrix model.","Inference: the explicit formula (1.7) can be checked numerically for small $N$ by simulating critical Ising interfaces on $\\delta\\mathbb{Z}^2$ and comparing the empirical probability that all interfaces hit the free arc with the predicted scaling limit.","Inference: carrying the same flow-line construction to $\\kappa=16/3$ would produce the analogous statement for FK-Ising interfaces or other alternating boundary patterns, with a different $\\beta=8/\\kappa$ value; the paper does not treat this case."],"forward_implications":["The two distinct flow-line constructions in Theorems 1.1 and 1.2 carry the same law, so any quantity of the multiple SLE—transition kernels, crossing events, endpoint laws—can be computed in whichever coupling is more convenient.","The hitting points of odd- and even-indexed curves have the explicit Selberg-type densities (1.3) and (1.4), and in the common-start limit they become intertwined $\\beta$-Jacobi ensembles with $\\beta=8/\\kappa$.","For the critical Ising model with alternating/free boundary, the probability that all $N$ interfaces end on the free boundary arc has the explicit conformal scaling limit (1.7) in terms of $Z_N$, $W_N$ and the Ising partition function $R_N$.","Conditioned on that event, the interface tuple $(\\gamma_1^\\delta,\\ldots,\\gamma_N^\\delta)$ converges under the curve metric to the $\\kappa=3$ member of the multiple SLE family.","The uniqueness theorem (Theorem 1.3) makes the characterization complete: any future construction satisfying the same conditional law describes the same object."],"supporting_citations":[{"why":"supplies the GFF flow-line theory: flow lines are SLE$_\\kappa(\\rho)$, their interactions, and the coupling that grounds Theorems 1.1, 1.2, 1.5 and 1.6.","marker":"[MS16a]"},{"why":"supplies the martingale and SLE coordinate-change formula used to derive the cascade relations and weighted laws in Lemmas 3.1 and 3.2.","marker":"[SW05]"},{"why":"provides the uniqueness-by-Markov-chain strategy for global multiple SLEs, including the lemma that fails in the present setting and must be modified.","marker":"[BPW21]"},{"why":"provides the Markov-chain uniqueness proof that the paper adapts, with the positivity estimate (3.20) as the key step.","marker":"[Zha24]"},{"why":"supplies the lemma on absolute continuity of SLE$_\\kappa((\\kappa-6)/2,(\\kappa-6)/2)$ flow lines on subdomains, used for Lemma 3.4 and (3.21).","marker":"[Yu23]"},{"why":"supplies convergence and the SDE for the Ising interface driving functions, and the nonzero/positivity properties of $R_N$ used in Theorem 1.9.","marker":"[Izy15]"},{"why":"supplies the explicit Ising partition functions $R_N$ and the multiple-interface driving SDE used in Lemma 5.1 and (1.7).","marker":"[FWY24]"},{"why":"supplies the commutation relations for SLEs behind the PDE system (5.10) satisfied by $R_N$.","marker":"[Dub07]"},{"why":"supplies the strong RSW estimate (Lemma 2.1) used to control the discrete interfaces in Lemma 5.2.","marker":"[CDCH16]"},{"why":"supplies the conjectured $\\beta=8/\\kappa$ relation between SLE-type curves and circular/Jacobi ensembles that the paper makes exact.","marker":"[Car03]"}],"fun_headline_variants":["Unique multiple SLE meets beta-Jacobi ensemble","SLE uniqueness unlocks Ising interface law","Critical Ising interfaces converge to SLE with Jacobi endpoints","Two SLE constructions, one law: beta-Jacobi connection"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the adapted Markov-chain uniqueness proof is valid—specifically that the positivity estimate (3.20) holds because the flow-line laws remain comparable when restricted to smaller domains; if either point fails, the two constructions could differ and the hitting-point and Ising conclusions would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Unique multiple SLE meets beta-Jacobi ensemble","SLE uniqueness unlocks Ising interface law","Critical Ising interfaces converge to SLE with Jacobi endpoints","Two SLE constructions, one law: beta-Jacobi connection"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000294,"raw_usage":{"total_tokens":1990,"prompt_tokens":1504,"completion_tokens":486,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":1120,"completion_tokens_details":{"reasoning_tokens":425}},"tokens_in":1120,"tokens_out":486,"duration_ms":4794,"temperature":1.0,"reasoning_tokens":425,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:45:50.333733+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For $N=2$, run the two couplings $P_2$ and $Q_2$ directly from the densities (1.3) and (1.4), simulate the GFF flow lines, and compare the distribution of the hitting point of the second curve: any difference would disprove the claimed equality of laws (Theorem 1.3). Alternatively, check positivity (3.20) for two tubes brought arbitrarily close to tangency; a zero transition probability would localize the failure in Lemma 3.4.","supporting_citations":[],"review_version":1}