{"id":"2332a8d8-b275-4f5a-a65c-4a80d757b146","arxiv_id":"2603.06119","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A weak solution of the anisotropic overdetermined torsion problem on an Ahlfors–David regular, weakly uniformly rectifiable set of finite perimeter exists if and only if the domain is a Wulff shape.","lead":"The authors prove that a weak solution to the anisotropic overdetermined torsion problem on a rough domain exists only when the domain is a Wulff shape. This extends Serrin-type rigidity from the Laplacian to anisotropic operators under mild geometric hypotheses that cover Lipschitz domains.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader correctly isolates the global β-square bound as the weakest geometric hypothesis and correctly notes that it is used only to control the intermediate-layer contribution in Lemma 2.2. That control is quantitative and standard (Proposition A.2), and the remainder of the argument (boundary blow-ups, Green-function comparison, P-function subharmonicity and equality case) follows classical elliptic/GMT lines once the volume identity is in hand. No stronger load-bearing concern surfaces upon re-examination of the full proof text. The verdict ACCEPT with high confidence is therefore left unchanged.","tokens_in":23538,"tokens_out":418,"duration_ms":4198,"concrete_test":"Independently re-derive the limit of R_{2}(ε) in Steps 6–9 of Lemma 2.3 starting from the algebraic identity of Lemma 2.2 and the measure representation (1.6), without invoking any Laplacian-specific cancellation; if the same identity (2.14) is recovered, the anisotropic adaptation is confirmed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of Theorem 1.1 is supported by a complete, self-contained argument. The volume identity (2.14) is obtained in Lemma 2.3 after the localization of Lemma 2.2, which uses the β-square bound (1.4) only through the covering Proposition A.2; that covering is standard under ADR and yields the required 1/|log r| gain on bad balls. The subsequent linearized Green-function representation (Lemma 3.1 and Remark 3.2) and the P-function rigidity (Section 3.3) close without further geometric hypotheses. The β-bound is already known for Lipschitz and uniformly rectifiable sets, so the theorem applies exactly where claimed. No hidden gap or circular step appears in the chain.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper proves an anisotropic Serrin rigidity theorem for rough domains: for a uniformly convex C^{2,γ} anisotropy H, a bounded indecomposable set of finite perimeter Ω admitting a distributional solution of the overdetermined anisotropic torsion problem (1.6) must, under Ahlfors–David regularity (1.5) and a global Jones β-square bound (1.4), be a translate and dilate of the Wulff body K, with explicit solution u=(r^{2}-H_*^{2})/(2n). The argument first establishes Lipschitz regularity and boundary blow-ups (Lemma 2.1), then a volume identity (2.14) via a β-controlled localization that replaces global W^{2,2} (Lemmas 2.2–2.3), and finally P-function rigidity for the linearized operator through a Green-function representation (Section 3).","tokens_in":23664,"tokens_out":838,"duration_ms":7258,"significance":"The result closes the anisotropic counterpart of the long-standing Lipschitz/rough-domain Serrin problem recently settled for the Laplacian in [FZ2025]. It applies directly to Lipschitz and uniformly rectifiable domains, where (1.4)–(1.5) hold, and supplies new analytic tools (Hessian smallness on interior balls, volume identity under only local W^{2,2}, anisotropic Green measure) that do not reduce to the isotropic case. The contribution is therefore both a genuine extension of classical overdetermined rigidity and a technical advance in the GMT treatment of anisotropic free-boundary problems.","major_comments":[],"minor_comments":[{"comment":"Abstract and Theorem 1.1: the abstract says “translate and dilation of the reflected Wulff shape -K” while the theorem statement says “homothetic to K”. Align the two formulations (and the introductory definition of PH) for consistency.","section":null},{"comment":"Lemma 2.1(5) and Step 6 of its proof: the anisotropic Liouville classification of half-space solutions is only sketched in a footnote. A short self-contained paragraph or a precise reference would help readers less familiar with the anisotropic setting.","section":null},{"comment":"Lemma 2.2, Step 7: the passage from the covering Proposition A.2 to the volume estimate (2.13) is dense; a one-sentence reminder that the intermediate region has thickness O(κε) and that each bad ball contributes volume ≲(κ^{-1}ε)^n would improve readability.","section":null},{"comment":"Section 3.1: the construction of the Green function G_x via exhaustion is standard, but the absolute continuity of L_A G_x+δ_x with respect to H^{n-1}⌊∂*Ω relies on the (n-1)-growth bound; a brief citation of the relevant potential-theory result would make the argument fully self-contained.","section":null},{"comment":"Typographical: “Kup to a translation” (p. 2) should be “K up to a translation”; a few other minor spacing issues appear around displayed equations in Section 2.","section":null}],"recommendation":"accept","confidential_remarks":"The manuscript is a natural and substantial sequel to the authors’ own [FZ2025]. The technical novelty is real and the proof chain is complete; I see no reason to delay acceptance. The journal’s readership in geometric analysis and free-boundary problems will find the result of clear interest."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is the anisotropic counterpart of Figalli–Zhang’s recent rough-domain Serrin theorem. Under ADR plus a global Jones β-square bound on an indecomposable set of finite perimeter, a weak solution of the anisotropic overdetermined torsion problem exists if and only if the domain is a translate/dilate of the Wulff body and the solution is the explicit quadratic. Lipschitz domains are included.\n\nWhat is actually new is not the high-level strategy (that follows their isotropic paper) but the replacements for the Laplacian-specific identities that fail for Δ_H. The load-bearing pieces are Lemma 2.2 (β-controlled localization that recovers the volume identity without global W^{2,2}) and the linearized Green-function representation plus maximum principle for H(∇u) that feed the P-function rigidity. Those steps are written carefully and appear self-contained; the appendix covering lemma that supplies the 1/|log r| gain on bad balls is standard under ADR.\n\nThe softest point is exactly the one the reader flags: the global β-square bound is essential for the localization near the boundary. It is not free, but it is a weak uniform-rectifiability hypothesis already known for Lipschitz and uniformly rectifiable sets, so the theorem applies where claimed. No circular reduction to the isotropic case, no free parameters, and the citation pattern is appropriate (their own prior work plus the classical anisotropic smooth results of Cianchi–Salani and Wang–Xia).\n\nThis is for people who work on overdetermined problems, free-boundary regularity, or anisotropic GMT. The proof is complete enough that a serious referee can check it line-by-line. I would send it out and I would cite it when I need the anisotropic rough-domain statement.","headline":"Solid anisotropic extension of the authors’ own rough-domain Serrin theorem; the new localization and linearized Green arguments close the proof under standard weak-UR hypotheses that cover Lipschitz domains.","tokens_in":24290,"tokens_out":446,"would_cite":true,"duration_ms":6103,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35N25","35J62"],"pacs":[],"model":"grok-4.5","headline":"Anisotropic overdetermined torsion forces rough domains to be Wulff shapes.","keywords":["overdetermined problems","anisotropic Laplacian","Wulff shape","sets of finite perimeter","Serrin rigidity","β-numbers","Ahlfors–David regularity"],"falsifier":"Exhibit a bounded indecomposable set of finite perimeter that is Ahlfors–David regular, admits a weak anisotropic overdetermined solution, yet is not homothetic to the Wulff body (or whose β-square integral diverges while a solution still exists).","tokens_in":24391,"feed_emoji":"◇","tokens_out":613,"duration_ms":5868,"temperature":0.7,"pith_summary":"Serrin’s classical theorem says that if a domain solves the overdetermined torsion problem for the ordinary Laplacian, it must be a ball. The same rigidity question for the anisotropic Laplacian—where surface tension is given by a general convex norm—had been settled only for smooth domains, and Laplacian-specific proofs did not transfer. This paper shows that the conclusion survives for far rougher domains: any bounded indecomposable set of finite perimeter whose reduced boundary is Ahlfors–David regular and satisfies a global Jones β-square bound admits a weak solution if and only if the domain is a translate and dilation of the Wulff body of the anisotropy. The solution is then unique and explicit. Lipschitz domains fall inside the hypotheses, so the result settles the anisotropic Serrin problem for every Lipschitz domain. The technical novelty is a localization argument that recovers a Weinberger-type volume identity without global second derivatives, using the β-bound to control the boundary layer.","feed_headline":"Rough domains that solve anisotropic torsion must be Wulff","feed_subtitle":"Lipschitz and uniformly rectifiable sets forced to the Wulff shape by a weak overdetermined condition","key_machinery":"The volume identity (n+2)∫_Ω u = c² n |Ω|, recovered by testing with dilation difference quotients and a β-controlled localization that replaces the missing global chain rule; once available, P-function subharmonicity forces equality and hence Wulff rigidity.","core_discovery":"Under Ahlfors–David regularity and a global β-number square-function bound on the reduced boundary, a distributional solution of the anisotropic overdetermined torsion problem exists if and only if, up to translation, the domain is homothetic to the Wulff body and the solution is the corresponding quadratic function of the dual norm.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Rough domains solve anisotropic torsion iff they are Wulff","Anisotropic overdetermined torsion forces domains to Wulff shape","Only Wulff bodies solve weak anisotropic Serrin for rectifiable sets","Anisotropic Serrin rigidity holds for Ahlfors-regular rough domains","Weak anisotropic torsion solutions exist solely for Wulff domains"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The global square-function bound on Jones β-numbers of the reduced boundary; without it the boundary-layer estimate needed for the volume identity fails.","fun_headline_variants_meta":{"raw":{"variants":["Rough domains solve anisotropic torsion iff they are Wulff","Anisotropic overdetermined torsion forces domains to Wulff shape","Only Wulff bodies solve weak anisotropic Serrin for rectifiable sets","Anisotropic Serrin rigidity holds for Ahlfors-regular rough domains","Weak anisotropic torsion solutions exist solely for Wulff domains"]},"model":"grok-4.5","effort":"low","cost_usd":0.005894,"raw_usage":{"total_tokens":1569,"prompt_tokens":784,"num_sources_used":0,"completion_tokens":89,"cost_in_usd_ticks":58940000,"prompt_tokens_details":{"text_tokens":784,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":696,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":784,"tokens_out":89,"duration_ms":6427,"temperature":1.0,"reasoning_tokens":696,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-15T14:00:21.835816+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Exhibit a bounded indecomposable set of finite perimeter that is Ahlfors–David regular, admits a weak anisotropic overdetermined solution, yet is not homothetic to the Wulff body (or whose β-square integral diverges while a solution still exists).","supporting_citations":[],"review_version":1}