{"id":"3c616bfb-19ca-4558-bc97-0b3c24d61279","arxiv_id":"math/0602442","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Any subset of a torus with two boundary components admits a proper holomorphic embedding into C^2, together with new sufficient conditions for bordered submanifolds of C^2.","lead":"This mathematics paper proves that certain bordered Riemann surfaces can be properly and holomorphically embedded into C^2. The main new case is any torus with two boundary components removed, which is not covered by earlier hyperelliptic methods.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The disc+point case of Theorem 2 rests on a local maximum-uniqueness assertion that is not actually proved: the announced inequalities leave the annulus Δ_r \\ Δ_{r/k} uncontrolled, and the half-plane used for 1/φ is not the one stated.","rationale":"The reader's weakest assumption correctly identified the local maximum-uniqueness assertion in Section 5 as the load-bearing step of Theorem 2. My reading of the proof agrees with that diagnosis and sharpens it: the displayed estimate (∗) does not by itself rule out larger values in the annulus Δ_r \\ Δ_{r/k}, and the subsequent argument about the minimum of |φ| and the injectivity of the square function is stated too tersely to verify the needed uniqueness. Because the theorem's statement explicitly includes boundary components that reduce to a point, this is not a marginal case: if the local assertion fails, the disc+point case is not proved and Theorem 2 loses its advertised scope. The concern is a proof gap rather than evidence of a false central claim, so the appropriate outcome is to accept the paper conditionally on a completed verification of the local maximum-uniqueness lemma, including control of the annulus and the correct half-plane for the reciprocal values. I do not see a flaw in the earlier Fatou-Bieberbach construction or in the two-disc case that is comparably load-bearing, and I credit the paper with a substantive new embedding result that is plausible and consistent with later developments.","tokens_in":9135,"tokens_out":25930,"duration_ms":258807,"concrete_test":"For a concrete square torus, fix a small disc D1 and a point q, choose z0 ∈ ∂D1 where Im z has a local minimum, set p_δ = z0 + iδ with δ = r/k, and numerically sample |\\wp(z − p_δ)| along the boundary curve γ_δ = ∂(φ_δ(D1)) for several small r and k. Check whether exactly one boundary point attains the maximum value and whether the maximizing value has a second boundary preimage. Simultaneously record whether all values 1/φ(z) for z ∈ γ_δ ∩ Δ_r lie in a single half-plane and whether the maximizer lies in Δ_{r/k}. If a second boundary point shares the same \\wp-value, the reduction to Theorem 1 collapses.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In the proof of Theorem 2, the case where D1 is a disc and D2 is a point q is reduced to Theorem 1 by claiming that for p_δ = z0 + iδ the maximum of |f_p| on ∂D1 occurs at a point with singleton fiber. That claim is the only bridge to Theorem 1, but the argument as written is incomplete. The estimate (∗) compares values inside Δ_{r/k} with values outside Δ_r; it does not control the annulus Δ_r \\ Δ_{r/k}, which is precisely where the arc γ_δ ∩ Δ_r can contain points with |z| close to r and hence potentially larger |\\wp|. The next assertion that 'φ restricted to γ_δ(r) ∩ Δ_r has to take its minimum at z0' is not proved: one must first show that the maximum of |\\wp| on the arc is attained in the interior rather than at the arc endpoints on |z| = r, and that the minimizing point z0 satisfies |z0| ≤ r/k so that (∗) applies. Finally, the uniqueness step invokes the square function being 1-1 on U_r, but the values to which it is applied are 1/φ(z) for z ∈ γ_δ ∩ Δ_r, which lie in the upper half-plane, not in the lower half-plane U_r; the proof only establishes φ(U_r) is contained in the lower half-plane. Without these missing controls, the singleton-fiber hypothesis of Theorem 1 is not established for the punctured boundary component, so the disc+point case of Theorem 2 is unsupported as written.","agreement_with_reader":"agree"},"referee_report":null,"author_rebuttal":null,"desk_editor":null,"pith_extraction":null,"created_at":"2026-08-28T20:13:05.510497+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}