{"id":"5135022c-7fa5-4e42-88cf-23a561e45879","arxiv_id":"2608.07381","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"For special toric degenerations of K3 surfaces, the intrinsic mirror and the universal toric degeneration mirror coincide after restricting to the minimal relative Gross-Siebert locus and basechanging by the polarization.","lead":"This paper proves that the intrinsic mirror construction agrees with the toric degeneration mirror construction for special K3 surface degenerations after an admissible resolution and a basechange. It unifies two major approaches within Gross-Siebert mirror symmetry and outlines the route toward higher-dimensional Calabi-Yau varieties.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.73 rests on an unverified extension of [J, Corollary 1.6] to ideals with non-maximal radical and to non-snc log structures; without this extension, Proposition 1.8 and the snc-dropping in Section 3.3.2 do not go through.","rationale":"The reader's CONDITIONAL verdict already identifies reliance on [J, Corollary 1.6] as the weakest assumption. My stress-test agrees with that identification but sharpens it: the load-bearing use is not merely the existence of birational invariance for finite-order mirrors over the maximal ideal, but its extension to arbitrary logarithmic modifications and to non-snc divisors. The paper's proof of Proposition 1.8 explicitly argues from the proof of [J, Corollary 1.6] rather than from its statement, and Section 3.3.2 replaces a technical proof with a series of 'one can check' assertions. These are internal flags that the paper itself acknowledges. Because admissible resolutions can introduce non-snc central fibres, the removal of the snc assumption is essential to the universal quantification in Conjecture 1.7. A positive resolution of the check would validate the proof; a negative one would leave the theorem incomplete. The final verdict remains CONDITIONAL because the claim may well be true and the missing verification is a hypothesis-check rather than a demonstrated counterexample.","tokens_in":67627,"tokens_out":7479,"duration_ms":67539,"concrete_test":"Obtain [J, Corollary 1.6] and its proof, and check the exact hypotheses: (a) does the statement or proof allow any monoid ideal I whose radical is a proper face K of P, not necessarily the maximal ideal; and (b) does it allow the divisorial log structure on X to be non-snc with non-globally-generated ghost sheaf? If either answer is no, consult the paper for an independent proof of that case. As a computational cross-check, use the explicit elliptic-curve mirrors of Chapter 2, perform a one-point log blowup that creates a contracted curve class, and verify directly from the theta-function products (2.3) that the extended mirror over the corresponding face K is the basechange predicted by Proposition 1.8; this tests the ideal-radical extension in a fully explicit setting.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central theorem is proved by first reducing to strongly admissible resolutions (Proposition 1.8) and by removing the simple normal crossings and global-generation hypotheses from the intrinsic mirror setup (Section 3.3.2). Both reductions are made to follow from [J, Corollary 1.6]. The paper itself flags the gap: Proposition 1.8 says 'the proof does not use the fact that the radical of I is the maximal ideal' and then extends the statement to ideals with radical J = P\\K for the classes of contracted curves. This is an inference from the proof of an external result, not a statement of that result. Similarly, Section 3.3.2 says the snc assumption 'can easily be removed' and 'one can check that all the arguments of [GS8] work' once [J] is available; these are assertions rather than proofs. Since admissible resolutions are defined as strongly admissible resolutions followed by arbitrary logarithmic modifications, the latter can have non-snc D, so Conjecture 1.7's 'for any admissible resolution' clause depends precisely on the non-snc birational invariance being asserted. If [J, Corollary 1.6] is only proved for ideals with maximal radical, or only for snc/globally-generated log schemes, then Proposition 1.8 is unsupported and Theorem 4.73 establishes Conjecture 1.7 only for strongly admissible resolutions, not for all admissible resolutions. This is not to say the claim is false; it is to say the central reduction is conditional on a stronger form of [J] than is quoted.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a comparison between two mirror constructions in the Gross-Siebert program: the algorithmic toric degeneration mirror and the intrinsic mirror. After an explicit treatment of degenerations of elliptic curves, the main result (Theorem A / Theorem 4.73) asserts that Conjecture 1.7 holds for special toric degenerations of K3 surfaces with smooth generic fibre: any such degeneration admits an admissible resolution to a minimal log Calabi-Yau degeneration, and the basechange of the intrinsic mirror by the polarization map is isomorphic to the toric degeneration mirror. The paper also proves a correspondence between the restriction of the intrinsic mirror to the minimal relative Gross-Siebert locus and the universal toric degeneration mirror (Theorem B / Theorem 5.35), and constructs log smooth resolutions for a natural family of toric degenerations of Calabi-Yau threefolds (Theorem C / Theorem 6.16). The argument is organized around reducing to strongly admissible resolutions, removing the simple normal crossings and global-generation hypotheses, and comparing the canonical and algorithmic scattering diagrams.","tokens_in":67900,"tokens_out":3326,"duration_ms":34738,"significance":"If Theorem 4.73 is correct, it establishes a major, long-sought compatibility between two central constructions in mirror symmetry, and the extension to the relative Gross-Siebert locus in Chapter 5 is a substantial additional result. The paper is unusually careful about stating its assumptions, and it provides explicit, checkable computations in the elliptic curve case and in the P^3 K3 example. The strength of the work is, however, conditional: the proof of the main theorem relies on an extension of birational invariance of punctured log Gromov-Witten invariants beyond the precise statement quoted from [J], and Section 3.3.2 contains assertions rather than proofs for the removal of the simple normal crossings assumption. Because these points are load-bearing for the reduction to strongly admissible resolutions, the central claim is currently established only up to this external input. The paper does not provide machine-checked proofs or reproducible code, but its explicit computations and clear structural organization are valuable regardless.","major_comments":[{"comment":"Proposition 1.8 is load-bearing for the reduction of Conjecture 1.7 to strongly admissible resolutions, yet its proof depends on an extension of [J, Corollary 1.6] that is not stated as a theorem anywhere. The sentence 'The proof does not use the fact that the radical of I is the maximal ideal' is an inference about the proof of an external result, not a proof of the needed statement. The needed statement concerns ideals I with radical J = P\\K for the curves contracted by the logarithmic modification, which is exactly the non-maximal-radical case. Without a supplied proof, or a precise citation of a theorem covering this case, Theorem 4.73 establishes Conjecture 1.7 only for strongly admissible resolutions with maximal-radical base ideals.","section":"§1, Proposition 1.8"},{"comment":"The removal of the simple normal crossings hypothesis is asserted rather than proved. The text states that the assumption 'can easily be removed' and that 'one can check that all the arguments of [GS8] work in this generalized setting', but no proof or detailed reference is given. This matters because admissible resolutions are allowed to have non-simple normal crossings D, so the comparison in Section 4.5 between the canonical scattering diagram and the algorithmic scattering diagram requires the non-snc version of the intrinsic mirror construction. The manuscript should either provide the promised argument or restrict Theorem 4.73 to the simple normal crossings case.","section":"§3.3.2"},{"comment":"The affine structure on the dual intersection complex in the general non-snc case is defined using [W, Theorem 4.1]. The hypotheses of that theorem are described only in passing, and it is not verified in the text that the log schemes X_{ρ_v} satisfy all of them in the situations needed for admissible resolutions. Since Construction 3.69 feeds directly into the definition of the canonical scattering diagram and hence into the main comparison, the precise content and applicability of [W, Theorem 4.1] should be stated and checked explicitly.","section":"§3.3.4, Construction 3.69"},{"comment":"The comparison of the canonical scattering diagram and the algorithmic scattering diagram for K3 surfaces relies on explicit punctured log Gromov-Witten computations taken from [G3] and [GHKS]. These computations are outsourced, and the text does not state exactly which results are used or verify that the hypotheses of those results are satisfied for the non-snc admissible resolutions introduced earlier. Since this comparison is the technical core of Theorem 4.73, the manuscript should list the precise external statements used and confirm that they apply in the full generality required.","section":"§4.5"}],"minor_comments":[{"comment":"The phrase 'Baryrev degeneration' appears once where 'Batyrev degeneration' is clearly intended; please fix the typo.","section":"§1, paragraph on Batyrev degenerations"},{"comment":"The notation X_{η(t)}^{∆} is introduced just before the proposition, but the symbol η(t) is reused for a power series without explicitly saying that it is the same function as in the displayed equation; a one-line clarification would help.","section":"§2.3, Proposition 2.6"},{"comment":"The proof of Proposition 3.63 invokes Steinitz's theorem via a PL-embedding of B into R^3, but B is only known to be a polyhedral manifold homeomorphic to a sphere; it would be helpful to state the precise version of Steinitz's theorem used and to note whether a triangulation step is required.","section":"§3.3.3, Construction 3.57"},{"comment":"The beginning of Section 3.4.1 is truncated in the displayed text after 'Restricting', and the section appears to be incomplete; please check that the final version contains the full subsection.","section":"§3.4.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a revised PhD thesis with a very broad scope. The main concern is not the novelty or the internal consistency but the dependence of the central theorem on an unverified extension of [J, Corollary 1.6] to non-maximal-radical ideals and non-snc log structures. If the author can supply a proof or a precise reference for that extension, or alternatively restrict the main theorems to the strongly admissible / snc setting, the manuscript would be much stronger. The outsourced scattering-diagram computations in Chapter 4 are a second point that the editor may wish to have checked independently."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper. First, it states and proves Conjecture 1.7 for special toric degenerations of K3 surfaces (Theorem 4.73), giving a broad correspondence between the intrinsic mirror and the toric degeneration mirror, and it extends that correspondence to a larger locus (Theorem 5.35) involving free parameters and gluing data. Second, the proof is conditional in a specific, flagged way: it depends on an extension of Johnston's birational invariance theorem [J, Cor. 1.6] to ideals with non-maximal radical and to non-snc log structures, and the paper does not prove that extension. It asserts that the proof of [J] goes through in that setting.\n\nWhat is genuinely new: Conjecture 1.7 is new, and Theorem A is the first proof of that conjecture in a nontrivial dimension. The comparison of algorithmic and canonical scattering diagrams is a real technical achievement, building on [GS3], [GS8], [GHK], and [AG]. The elliptic curve chapter is a nice explicit warm-up with concrete equations. The paper is carefully structured, defines \"special\" and \"admissible resolutions\" precisely, and is honest about what is assumed. It explicitly says \"the proof does not use...\" and \"one can check that all arguments work\" at the two spots where it extends [J]. That transparency is to its credit.\n\nThe soft spot is real and load-bearing. Proposition 1.8 reduces Conjecture 1.7 to strongly admissible resolutions; Section 3.3.2 removes the snc and global-generation assumptions. Both rely on an unproved strengthening of [J]. If [J] only holds for maximal radical or snc, then Theorem 4.73 proves Conjecture 1.7 only for strongly admissible resolutions, not all admissible resolutions. The paper flags this but does not close it. Also, the K3 scattering computations are outsourced to [G3] and [GHKS]; I cannot verify them, and a referee will need to. The higher-dimensional part is mostly conjectural and conditional on Conjecture 6.21, which is fine if framed as such.\n\nOverall, the central argument appears coherent, there is no circularity, and the external dependencies are explicit. The paper deserves a serious referee, but the referee should press on the [J] extension before accepting Theorem A in full generality. For a colleague working on Gross-Siebert mirror symmetry, this is a citation-worthy result, with a caveat. I would bring the elliptic curve chapter to a reading group, and the main theorem if the group has log Gromov-Witten background.\n\nRecommendation: send to peer review, and require the author to either prove the needed extension of [J] or state the theorem with the reduced hypothesis explicitly built in.","headline":"Strong thesis proving the K3 case of a new Gross-Siebert mirror conjecture, but the proof's load-bearing reduction rests on an unproved extension of Johnston's birational invariance theorem.","tokens_in":68498,"tokens_out":3016,"would_cite":true,"duration_ms":26369,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14J33","14J28","14N35","14M25","14T05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the intrinsic mirror construction, after resolving a special toric degeneration of K3 surfaces and base-changing by the polarization, yields a family isomorphic to the classical toric degeneration mirror.","keywords":["mirror symmetry","toric degenerations","intrinsic mirror symmetry","K3 surfaces","scattering diagrams","log Calabi-Yau degenerations","punctured log Gromov-Witten invariants","log smooth resolutions"],"falsifier":"Compute the two mirror families, or their defining scattering diagrams, modulo $t^2$ for a specific special toric degeneration of K3 surfaces such as the quartic hypersurface in $\\mathbb P^3$ degenerating to the coordinate tetrahedron, and check whether the basechanged intrinsic mirror is isomorphic to the toric degeneration mirror; a mismatch in any leading-order wall or equation would disprove Theorem A.","tokens_in":67346,"feed_emoji":"🪞","tokens_out":7791,"duration_ms":63974,"temperature":0.7,"pith_summary":"Mirror symmetry has two competing constructions for Calabi-Yau families with toric special fibres: an older algorithmic one that reconstructs a mirror from a scattering diagram built out of the combinatorial data of the degeneration, and a newer intrinsic one defined through punctured log Gromov-Witten invariants. This paper shows they agree for K3 surfaces. For any special toric degeneration of K3 surfaces with a polarization, the author constructs an admissible resolution to a minimal log Calabi-Yau degeneration and proves that the intrinsic mirror of the resolution, base-changed along the map sending a curve class to its intersection with the pulled-back polarization, is isomorphic to the toric degeneration mirror. The comparison is made by relating the two scattering diagrams that encode the two constructions. The paper also extends the correspondence to the minimal relative locus of the extended intrinsic mirror, matching the universal toric degeneration mirror, and constructs log smooth resolutions for a natural family of toric degenerations of Calabi-Yau threefolds.","feed_headline":"Two mirror constructions coincide for K3 surfaces","feed_subtitle":"A proof shows the intrinsic mirror recovers the classical toric degeneration mirror after resolving the family.","key_machinery":"The load-bearing object is the scattering diagram: a finite set of walls, each carrying a function, on the dual intersection complex of a degeneration, encoding the curve-counting corrections that define a mirror family. The paper compares the algorithmic scattering diagram $\\bar{\\mathfrak D}$ of a toric degeneration with the canonical scattering diagram $\\mathfrak D$ of its admissible log-smooth resolution. The canonical diagram is defined from wall types and punctured log Gromov-Witten invariants; the comparison, after the base change $h$, shows the two diagrams are combinatorially equivalent, and by the uniqueness property of the algorithmic construction this forces the two mirror families to be isomorphic. A second key device is the admissible resolution itself: blowing up components of the central fibre to make the degeneration log smooth and minimal log Calabi-Yau while keeping control of the dual intersection complex and curve classes.","core_discovery":"The central discovery is Theorem A: Conjecture 1.7 holds for special toric degenerations of K3 surfaces. Concretely, every such degeneration $\\bar{\\mathfrak X}\\to \\mathcal S$ with polarization $A$ admits an admissible resolution $\\pi:\\mathfrak X\\to \\bar{\\mathfrak X}$ to a minimal log Calabi-Yau degeneration, and the basechange of the intrinsic mirror $\\check{\\mathfrak X}\\to \\operatorname{Spf} \\widehat{k[P]}$ by the homomorphism $h:\\beta\\mapsto \\pi_* A\\cdot \\beta$ is isomorphic to the toric degeneration mirror $\\check{\\bar{\\mathfrak X}}\\to \\operatorname{Spec} k[t]$. The proof constructs the resolution, interprets the extended intrinsic mirror through scattering diagrams, and proves that the canonical scattering diagram of the intrinsic construction becomes combinatorially equivalent to the algorithmic scattering diagram of the toric degeneration mirror after the base change. In addition, Theorem B upgrades this to an isomorphism between the restriction of the intrinsic mirror to the minimal relative Gross-Siebert locus and a subfamily of the universal toric degeneration mirror, and Theorem C provides strongly admissible resolutions for a class of toric degenerations of Calabi-Yau threefolds obtained by the reconstruction algorithm.","pith_inferences":["If the comparison extends to higher dimensions, the intrinsic mirror would supply a canonical choice of the initial slab functions and gluing data that the algorithmic toric degeneration mirror currently requires as input, removing a source of non-uniqueness.","The reliance on birational invariance suggests the intrinsic mirror should be independent of the chosen admissible resolution; if true, it would be a canonical mirror attached to the original degeneration rather than to a resolution.","A concrete testable extension would be to carry out the scattering-diagram comparison explicitly for the quartic K3 in $\\mathbb P^3$ and verify the basechanged canonical diagram equals the algorithmic one wall by wall through all orders in $t$."],"forward_implications":["For K3 surfaces, the intrinsic mirror construction is no longer a separate object: it recovers the toric degeneration mirror exactly, so results and algorithms for one construction transfer to the other.","The toric degeneration mirror is shown to depend only on the intended data — central fibre, polarization, and chosen slab data — by being identified with the intrinsic mirror after base change.","The extended correspondence over the minimal relative Gross-Siebert locus means the intrinsic mirror contains the universal family of the toric degeneration mirror, varying freely in initial slab functions and gluing data.","In dimension three, a natural class of toric degenerations admits, after a bounded base change, strongly admissible resolutions, giving a path toward the same comparison for Calabi-Yau threefolds.","For elliptic curves, the two mirror constructions agree tautologically and the universal toric degeneration mirror is isomorphic to the intrinsic mirror, giving the base case of the general picture."],"supporting_citations":[{"why":"It supplies the reconstruction algorithm that builds the toric degeneration mirror from the algorithmic scattering diagram.","marker":"[GS3]"},{"why":"It provides the original definition of the intrinsic mirror via three-pointed punctured stable maps.","marker":"[GS7]"},{"why":"It gives the canonical scattering diagram interpretation of the intrinsic mirror and its equivalence with the theta-function mirror.","marker":"[GS8]"},{"why":"It supplies the universal toric degeneration mirror used in Theorem B and the gluing conventions adopted throughout.","marker":"[GHS]"},{"why":"It supplies the birational invariance of punctured log Gromov-Witten invariants used in Proposition 1.8 to reduce to strongly admissible resolutions.","marker":"[J]"},{"why":"It provides the punctured log Gromov-Witten theory that defines the wall types and structure constants.","marker":"[ACGS2]"},{"why":"It provides the finite scattering diagram for the extended intrinsic mirror in the basic K3 case, which the paper generalises.","marker":"[GHKS]"},{"why":"It performs explicit scattering diagram computations for K3 degenerations that feed into the comparison in Section 4.5.","marker":"[G3]"},{"why":"It defines divisorial log deformations and their local models, used to identify special toric degenerations of K3 surfaces.","marker":"[GS2]"}],"fun_headline_variants":["Intrinsic mirror matches toric degeneration mirror for K3s","K3 mirrors unified: intrinsic and toric degeneration agree","Theorem: Intrinsic mirror recovers toric degeneration mirror","Mirror symmetry for K3 surfaces: two constructions coincide","Intrinsic mirror equals toric degeneration mirror for K3"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument leans on an externally supplied result, birational invariance of punctured log Gromov-Witten invariants, which is used to reduce Conjecture 1.7 to strongly admissible resolutions and to remove a technical global-generation assumption; if that invariance fails, the reduction and the scattering-diagram comparison do not go through.","fun_headline_variants_meta":{"raw":{"variants":["Intrinsic mirror matches toric degeneration mirror for K3s","K3 mirrors unified: intrinsic and toric degeneration agree","Theorem: Intrinsic mirror recovers toric degeneration mirror","Mirror symmetry for K3 surfaces: two constructions coincide","Intrinsic mirror equals toric degeneration mirror for K3"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000517,"raw_usage":{"total_tokens":2570,"prompt_tokens":1070,"completion_tokens":1500,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":686,"completion_tokens_details":{"reasoning_tokens":1418}},"tokens_in":686,"tokens_out":1500,"duration_ms":10503,"temperature":1.0,"reasoning_tokens":1418,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:27:36.054105+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the two mirror families, or their defining scattering diagrams, modulo $t^2$ for a specific special toric degeneration of K3 surfaces such as the quartic hypersurface in $\\mathbb P^3$ degenerating to the coordinate tetrahedron, and check whether the basechanged intrinsic mirror is isomorphic to the toric degeneration mirror; a mismatch in any leading-order wall or equation would disprove Theorem A.","supporting_citations":[],"review_version":2}