{"id":"7641a3d7-15f1-4713-976a-9e7653998ffd","arxiv_id":"2412.10698","paper_version":4,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"The paper claims, via mirror symmetry and a quiver search, that none of a listed set of cE6 and cE7 singularities admit crepant resolutions and hence lack N=1 quiver SCFT duals.","lead":"The paper uses mirror symmetry to argue that most weighted homogeneous Gorenstein cDV singularities in two cE families have no crepant resolution, so they cannot support the usual four-dimensional superconformal quiver gauge theories from D3-branes. A generalist might care because it offers a new way to decide which singular geometries can serve as holographic gravity duals, though the main conclusion depends on unproved conjectures.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No-go claim rests on the unproved 'only if' direction of the Evans-Lekili conjecture (conceded in §2.3); the physics search in §4 is constrained, so the abstract's 'demonstrate/confirmed' overstates what is established.","rationale":"I agree with the reader that the central inference is conditional on the unproved 'only if' direction of the Evans-Lekili conjecture and that the abstract overclaims. The paper deserves credit for transparency: §2.3 explicitly flags the assumption, and the k=8 (cE6) computation in §3.1 includes a genuine analytic period-4 argument covering all negative degrees. Nevertheless, the decisive step converting Tables 3 and 4 into non-existence statements is the contrapositive of a conjecture proved only for cA_n ([35]); neither the constrained §4 search nor the §5 'compelling evidence' phrasing supplies a proof. I evaluated the secondary truncation concern and found it is not load-bearing for the no-go cases: within the displayed range the negative-degree ranks already take multiple values (e.g., Table 3, k=2: ranks 3 and 2; Table 4, k=4: ranks 2 and 1), so constant rank is already excluded regardless of behavior below d=-10; the truncation only weakens the 'stabilizes iff k=8/2/14' formulation. The more serious compounding issue, underemphasized by the reader, is the self-flagged non-isolated fixed subspace in §3.2: if the simplification of (2.11) breaks down for γ fixing y and z but not w, the cE7 ranks and the HH²=0 input to (2.9) are unverified, and the entire cE7 half of the claim inherits that gap. The proposed cD_n cross-check is decisive because [6] supplies independent resolution statuses for a family where the conjecture direction is unproved and [35] supplies a direct computational route; one mismatch refutes the inference, while full consistency converts the concern into a clearly stated conditionality. On the verdict, I recommend no change: the advertised demonstration is conjecture-conditional, the confirmation search is constrained, and the §3.2 caveat is unresolved, so the reader's REJECT stands; a CONDITIONAL pass would be defensible only if the abstract were rewritten to state the conjecture dependence and the §3.2 caveat were resolved.","tokens_in":29290,"tokens_out":41492,"duration_ms":330786,"concrete_test":"Test the unproved 'only if' direction of the Evans-Lekili conjecture on a cD_n singularity from Table 1 whose crepant-resolution status is fixed independently by [6] (the authors' cited source for Claim 1.1), computing its negative-degree symplectic cohomology with the direct method of [35] — the technique that already proved the conjecture for cA_n. If a cD_n singularity known to admit no crepant resolution nevertheless has constant negative ranks equal to f(X), or a resolvable one has non-constant ranks, the 'only if' direction fails exactly in the regime where the paper applies it, and the inference from Tables 3 and 4 fails. If every such cD_n case is consistent, the load-bearing assumption is corroborated where it is currently unproved, and the paper's residual defect becomes an explicitly stated proof obligation rather than a demonstrated result.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central result — that no singularity in Problem 2.1 admits a crepant resolution, which is the 'only if' part of Claim 1.1 — is not established by the paper's method; it is conditional on the 'only if' direction of the Evans-Lekili conjecture ([28], §2.3), namely: a cDV singularity admitting a crepant resolution with l irreducible exceptional curves must have negative-degree symplectic cohomology of constant rank l. The paper states this plainly: \"Under the assumption that the 'only if' direction of the above conjecture holds, this result leads to the main claim of our paper in the introduction.\" The inference from the non-constant ranks in Tables 3 and 4 to absence of a crepant resolution is precisely the contrapositive of that conjecture direction, which is proved only for cA_n ([35]). The §4 physics verification does not close the gap: it is an explicitly constrained enumeration (R-charges in [0,2], charges in multiples of 1/m, paired bifundamentals with equal charges), so it cannot exclude duals outside the search space, and the abstract's 'confirmed by enumerating all consistent quiver gauge theories' overstates a search the body calls 'subject to additional constraints.' A compounding technical gap is self-flagged in §3.2: when γ fixes y and z but not w, the restricted singularity Ẇ_γ is not isolated, so the simplification of (2.11) on which the A/B/C-monomial counting rests 'may break down,' with no argument that the caveat is immaterial; if it does, Table 4 ranks, including the HH²=0 required for (2.9), could change. The d=-10 truncation is not itself the obstacle for the no-go cases, since the displayed ranks already take multiple values within the range and constancy is already excluded, but the asserted 'stabilizes only if k=8 (resp. k=2,14)' is an extrapolation from finite tables. The claim may be true (the authors call it a corollary of [6]), but the advertised demonstration is conjecture-dependent.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies isolated weighted homogeneous Gorenstein compound Du Val (cDV) singularities of cE6 and cE7 type in the K-stable range, and claims that none of the singularities listed in Problem 2.1 admit a crepant resolution; the only exceptions are four known singularities (Claim 1.1). The method is homological: the negative-degree symplectic cohomology of the Milnor fiber is identified, via homological mirror symmetry, with the Hochschild cohomology of equivariant matrix factorizations of the Berglund–Hübsch mirror, and the Evans–Lekili criterion is used to convert constant negative-degree rank ℓ into existence of a crepant resolution with ℓ exceptional curves. The rank computations are worked out for the cE6 and cE7 families and are tabulated for degrees −10 ≤ d ≤ 2, with k = 8 and k = 2, 14 identified as the only cases whose ranks appear to stabilize. Section 4 attempts a field-theoretic cross-check by enumerating quiver gauge theories with matching Hilbert series. The paper is clearly written and contains a detailed worked example, but its main no-go conclusion is explicitly conditional on an unproved conjecture and on finite truncation of the degree range.","tokens_in":29587,"tokens_out":6065,"duration_ms":58012,"significance":"If fully established, the paper would resolve an open question about crepant resolutions of a family of K-stable cDV singularities and would rule out 4d N=1 superconformal quiver duals for those geometries, with direct consequences for AdS/CFT applications. The paper has genuine strengths: the translation of the symplectic-cohomology computation into finite linear congruences is explicit, the k=8 example is worked out in detail, and the attached Mathematica notebook makes the enumeration reproducible. However, the central non-existence claim is not proved as stated: it relies on the unproved 'only if' direction of the Evans–Lekili conjecture, the stabilization claims are read off from tables truncated at d = −10, and the physics-side 'confirmation' is a constrained search rather than an exhaustive enumeration. The conditional computational framework is interesting, but the significance of the unconditional claims is not currently supported.","major_comments":[{"comment":"The main no-existence conclusion is conditional on the unproved 'only if' direction of the Evans–Lekili conjecture. The paper states this explicitly: 'Under the assumption that the “only if” direction of the above conjecture holds, this result leads to the main claim of our paper in the introduction.' Since the inference from nonconstant negative-degree symplectic cohomology ranks to absence of a crepant resolution is precisely the contrapositive of that conjecture direction, and since the conjecture is proved only for cA_n singularities, Claim 1.1 and the abstract's 'demonstrate' overstate what is established. The paper should either prove or cite a proof of the missing direction, or present the main result as explicitly conditional.","section":"§2.3 and Claim 1.1"},{"comment":"The cE7 computation contains a self-flagged gap: the text notes that if γ fixes y and z but not w, then the restricted singularity Ẇ_γ is not isolated and 'the simplification below (2.11) may break down,' with only the expectation that the caveat is immaterial. The simplified counting of A/B/C monomials relies on the cohomology of the Koszul complex being concentrated in low degrees with a Jacobian-ring basis, and this is exactly what can fail for non-isolated Ẇ_γ. Since Table 4 and the resulting no-crepant-resolution conclusion depend on those counts, an argument showing that the non-isolated cases do not affect negative-degree ranks or HH^2=0 is needed.","section":"§3.2"},{"comment":"The stabilization statements are inferred from finite tables: ranks are listed only for −10 ≤ d ≤ 2, and the conclusion that rank stabilizes only for k=8 (cE6) and k=2,14 (cE7) is read off from that finite window. The argument requires knowledge of negative-degree symplectic cohomology for all d < 0; a sequence that is nonconstant down to d = −10 could in principle become constant at d < −10. No generating-function expression, closed-form recurrence, or upper bound on the vanishing range is supplied. The assertion that stabilization occurs only in the listed cases is therefore not supported by the data presented.","section":"§3.1–3.2, Tables 3 and 4"},{"comment":"The physics-side verification is not an enumeration of all consistent quiver gauge theories. The search imposes three additional constraints: R-charges in [0,2], equal R-charges for each pair of bifundamentals, and R-charges restricted to multiples of 1/m. The body of §5 acknowledges that 'the search is subject to additional constraints,' but the abstract says the absence of duals is 'confirmed by enumerating all consistent quiver gauge theories.' These constraints are introduced for simplicity and are not derived from consistency, so the search cannot exclude superconformal quivers outside the restricted class. The abstract and Claim 1.1 should be softened, or the constraints should be justified as consequences of consistency.","section":"§4.2 and Abstract"}],"minor_comments":[{"comment":"The introduction calls the Evans–Lekili statement an 'important conjecture established in [28],' while §2.3 correctly describes it as a conjecture whose 'only if' direction is unproved except for cA_n; this inconsistency should be fixed.","section":"§1 and §2.3"},{"comment":"The footnote says 'The non-isolated singularities considered in this paper admit at least two C* actions,' but the paper explicitly restricts to isolated cDV singularities; either the footnote refers to examples from [13] or it is misworded.","section":"§5, footnote 16"},{"comment":"There are numerous typos and infelicities: 'accosiated', 'Berlund-Hübsch', 'stablizes', 'T able', 'Pinckham', 'kernal', and 'mulitplication' are examples. A careful proofreading pass is needed.","section":"Throughout"},{"comment":"The Futaki-invariant formula (4.4) is stated for test configurations generated by coordinate vector fields, but the paper does not discuss whether these exhaust the relevant test configurations for K-stability; if not, the K-stable range in Table 1 should be cited as an external result rather than derived from formula (4.4) alone.","section":"§4.1"}],"recommendation":"reject","confidential_remarks":"The concerns raised in review are accurate: the central non-existence theorem is conditional on an unproved conjecture, the stabilization claims rely on truncated tables, and the physics search is explicitly constrained. These are load-bearing for the paper's main claim. The underlying computational method and the worked examples may be worth developing further, but as it stands the manuscript does not establish the announced result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's real content is a Hochschild cohomology computation for two families of cDV singularities via Berglund–Hübsch–Krawitz mirror symmetry, together with a constrained quiver Hilbert series search. That part is honestly done and new as an application. But the main claim is not new—Claim 1.1 is explicitly a corollary of [6]—and the paper's own argument for non-existence is conditional on the 'only if' direction of the Evans–Lekili conjecture, which is proved only for cA_n. The authors say this plainly in §2.3: 'Under the assumption that the “only if” direction of the above conjecture holds, this result leads to the main claim.' So the abstract's 'demonstrate' overstates what is established.\n\nWhat the paper does well: the explicit HH* tables for these cE6/cE7 mirrors are a genuine service; I have not seen them worked out to this degree. The Mathematica implementation is the kind of reproducible evidence that helps. The Morrison–Pinkham check in §4 is a good cross-validation, and the prose is unusually clear about which steps are conjectural and which are computational. That honesty deserves credit.\n\nSoft spots, in order of importance. First, the load-bearing inference from non-constant ranks to absence of a crepant resolution uses precisely the unproved 'only if' direction of the conjecture. Without that, the paper is a computation, not a proof. Second, the paper itself flags in §3.2 that when γ fixes y and z but not w, the restricted singularity Ẇ_γ is not isolated and the simplification of (2.11) 'may break down'; the caveat is then called 'expected immaterial' with no argument. This is a real gap in the cE7 computation, and if it materializes, the Table 4 ranks—including the HH^2=0 needed for (2.9)—could change. Third, the d=-10 truncation matters less for the no-go cases than for the four 'yes' cases: for k≠8 the ranks already vary within the table, which is enough to exclude constancy, but for k=8,2,14 the claim that ranks stabilize at all negative degrees is an extrapolation from finite data with no analytic control below -10. Fourth, the physics search in §4 is not an exhaustive enumeration of all consistent quiver theories; it restricts R-charges to [0,2], to multiples of 1/m, and to equal charges for paired bifundamentals. The body is honest about this ('subject to additional constraints'), but the abstract's 'confirmed by enumerating all consistent quiver gauge theories' goes beyond what the search actually does.\n\nThe paper is for readers working on holographic duals of cDV singularities. It will not replace [6] as the source of the classification, and the mirror-theoretic route remains conjecture-dependent. Still, it is a serious, transparent computation with reproducible data and a clearly stated dependence on an open conjecture. I would send it to a referee, expecting the referee to ask for a proof or removal of the §3.2 caveat, an analytic statement about stabilization beyond -10, and an abstract that matches the body. That is a major revision, not a desk reject.","headline":"A careful mirror-symmetry computation whose no-go conclusion rests on an unproved conjecture the authors explicitly concede; worth a referee, but the abstract oversells it.","tokens_in":30275,"tokens_out":3125,"would_cite":false,"duration_ms":29884,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14J17","53D37","16E40","14E15","81T60"],"pacs":[],"model":"deepseek-v4-flash","headline":"A mirror-symmetry computation rules out crepant resolutions for 24 cDV singularities.","keywords":["compound Du Val singularities","crepant resolution","symplectic cohomology","homological mirror symmetry","Hochschild cohomology","matrix factorizations","N=1 quiver gauge theory","AdS/CFT correspondence"],"falsifier":"An explicit crepant resolution of any one of the 24 singularities in Problem 2.1, for example a small resolution of $x^2+y^3+z^4+y w^5$ with one exceptional curve, would falsify the main claim. Short of that, extending the Hochschild computations in Tables 3 and 4 to degrees below $-10$ and finding a previously excluded $k$ whose negative-degree ranks become constant would break the stabilization argument.","tokens_in":29038,"feed_emoji":"🚫","tokens_out":15600,"duration_ms":123913,"temperature":0.7,"pith_summary":"The paper asks which weighted homogeneous Gorenstein compound Du Val (cDV) threefold singularities admit crepant resolutions, and therefore which can serve as backgrounds for 4d $\\mathcal{N}=1$ superconformal quiver gauge theories on D3-branes. It argues that among the K-stable $E_6$ and $E_7$ families, only four types up to weight-one deformations are resolvable, and that none of the remaining unresolved cases listed in Problem 2.1 admit a crepant resolution. The route is indirect: compute a symplectic invariant of the Milnor fiber, the negative-degree symplectic cohomology, through homological mirror symmetry and Hochschild cohomology of equivariant matrix factorizations. A companion enumeration of all consistent two- and three-node quiver gauge theories finds no candidate dual, matching the geometric no-go result. The geometric conclusion is stated under the assumption of the unproved 'only if' direction of the conjecture of [28].","feed_headline":"24 singularities ruled out as holographic backgrounds","feed_subtitle":"Mirror-symmetry computation finds no crepant resolution for any of them, so none can support a 4d N=1 superconformal quiver dual.","key_machinery":"The load-bearing object is the negative-degree symplectic cohomology $SH^{<0}(M_X)$ of the Milnor fiber of the singularity; the criterion from [28] says a crepant resolution with $l$ irreducible exceptional curves forces this invariant to have constant rank $l$ in every negative degree. Since direct computation is hard, the paper computes it on the mirror side: homological mirror symmetry identifies $SH^\\bullet$ with the Hochschild cohomology $HH^\\bullet$ of the category of equivariant matrix factorizations of the mirror superpotential, and the formula of [34] reduces each rank to counting solutions of explicit integral linear congruences associated to group elements fixing coordinate subspaces. Constant rank across all negative degrees is read off from tables; nonconstant rank is the no-go signal.","core_discovery":"The central claim, stated on the paper's own terms, is a dichotomy for the K-stable weighted homogeneous cDV singularities with $J=E_n$ in Table 1: up to weight-one deformations, exactly four equations admit a crepant resolution, namely $x^2+y^3+z^4+w^{12}=0$, $x^2+y^3+yz^3+w^{18}=0$, $x^2+y^3+z^5+w^{30}=0$, and $x^2+y^3+yz^3+w^{2}z=0$. Consequently, none of the 24 singularities collected in Problem 2.1 admit a crepant resolution, and according to the AdS/CFT dictionary they cannot be holographic backgrounds for a 4d $\\mathcal{N}=1$ superconformal quiver gauge theory. The argument runs through the symplectic-cohomology criterion of [28]: after mirror symmetry, the negative-degree ranks are computed as Hochschild cohomology ranks of equivariant matrix factorizations, and they are constant across all negative degrees exactly in the four known resolvable cases. The paper verifies the conclusion from the field-theory side by enumerating all consistent two- and three-node quiver gauge theories and finding none with the correct Hilbert series.","pith_inferences":["If the 'only if' conjecture from [28] is true in full, the stabilization computation becomes a general finite arithmetic test for crepant resolutions of invertible cDV singularities, no resolution construction required.","The tables only probe degrees down to $-10$; proving a lower bound on the degrees where new contributions can appear, or extending the search much deeper, would close the residual arithmetic loophole in the stabilization claim.","The candidate quiver dualities found in the last section suggest that matching quiver Hilbert series can generate new gauge-theory dual pairs; comparing their large-N superconformal indices is a concrete check the paper leaves open.","The no-go statement targets quiver gauge theories; it does not rule out other 4d $\\mathcal{N}=1$ SCFTs, such as non-Lagrangian or non-quiver theories, as holographic duals of these singularities."],"forward_implications":["The 24 singularities in Problem 2.1 cannot serve as the transverse geometry for D3-branes in a 4d $\\mathcal{N}=1$ superconformal quiver gauge theory, under the standard holographic dictionary.","Within the K-stable range, the only $E_n$-type singularities in the paper's list that admit a crepant resolution are the four displayed equations, up to weight-one deformations.","Constant negative-degree rank is a workable sufficient signal for crepant resolvability, and nonconstant rank is an obstruction, so the same Hochschild-side computation can be reused on other invertible cDV families.","The exhaustive quiver search finds no two- or three-node candidate with matching Hilbert series, so the field-theoretic side independently reproduces the geometric no-go."],"supporting_citations":[{"why":"Supplies the conjecture linking negative-degree symplectic cohomology rank to the number of exceptional curves in a crepant resolution, and the isomorphism to Hochschild cohomology used for the computation.","marker":"[28]"},{"why":"Provides the Hochschild cohomology formula for equivariant matrix factorizations that reduces the computation to counting integral solutions.","marker":"[34]"},{"why":"Gives the cA_n proof of the symplectic-cohomology criterion and the stabilization method that this paper generalizes to the cE families.","marker":"[35]"},{"why":"Prior physics-based determination of crepant resolutions for the singularities in Table 1; Claim 1.1 is described as a corollary of its results.","marker":"[6]"},{"why":"Establishes that the dual quiver gauge theory is determined by a noncommutative crepant resolution, connecting the geometry to the field theory.","marker":"[12]"},{"why":"Shows existence of noncommutative crepant resolutions is equivalent to existence of crepant resolutions for isolated cDV singularities.","marker":"[14]"},{"why":"Classifies the possible dual graphs of exceptional curves for small resolutions, restricting the quiver topologies searched in Section 4.","marker":"[50]"},{"why":"Provides the mass-deformation counts f(X) recorded in Table 2, used to identify which singularities could have nontrivial exceptional curves.","marker":"[33]"}],"fun_headline_variants":["24 cDV singularities fail AdS/CFT test","Mirror symmetry excludes 24 holo duals","No crepant resolution: 24 singularities no-go","Holographic no-go for 24 quiver duals","24 singularities cannot host 4d N=1 quivers"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the unproved 'only if' direction of the conjecture of [28]: a cDV singularity with a crepant resolution containing $l$ irreducible exceptional curves must have negative-degree symplectic cohomology of constant rank $l$, and a counterexample to that direction would remove the force of the no-go conclusion.","fun_headline_variants_meta":{"raw":{"variants":["24 cDV singularities fail AdS/CFT test","Mirror symmetry excludes 24 holo duals","No crepant resolution: 24 singularities no-go","Holographic no-go for 24 quiver duals","24 singularities cannot host 4d N=1 quivers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000657,"raw_usage":{"total_tokens":2984,"prompt_tokens":901,"completion_tokens":2083,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":517,"completion_tokens_details":{"reasoning_tokens":1998}},"tokens_in":517,"tokens_out":2083,"duration_ms":14174,"temperature":1.0,"reasoning_tokens":1998,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:41:36.945680+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"An explicit crepant resolution of any one of the 24 singularities in Problem 2.1, for example a small resolution of $x^2+y^3+z^4+y w^5$ with one exceptional curve, would falsify the main claim. Short of that, extending the Hochschild computations in Tables 3 and 4 to degrees below $-10$ and finding a previously excluded $k$ whose negative-degree ranks become constant would break the stabilization argument.","supporting_citations":[{"cited_title":"Symplectic cohomology of compound Du Val singularities","cited_arxiv_id":"2104.11713","evidence_quote":"Supplies the conjecture linking negative-degree symplectic cohomology rank to the number of exceptional curves in a crepant resolution, and the isomorphism to Hochschild cohomology used for the computation."},{"cited_title":"Homological mirror symmetry for Milnor fibers of simple singularities","cited_arxiv_id":"2004.07374","evidence_quote":"Provides the Hochschild cohomology formula for equivariant matrix factorizations that reduces the computation to counting integral solutions."},{"cited_title":"Symplectic cohomology of quasihomogeneous $cA_n$ singularities","cited_arxiv_id":"2404.17301","evidence_quote":"Gives the cA_n proof of the symplectic-cohomology criterion and the stabilization method that this paper generalizes to the cE families."},{"cited_title":"De Marco, A","cited_arxiv_id":null,"evidence_quote":"Prior physics-based determination of crepant resolutions for the singularities in Table 1; Claim 1.1 is described as a corollary of its results."},{"cited_title":"Aspinwall and D.R","cited_arxiv_id":null,"evidence_quote":"Establishes that the dual quiver gauge theory is determined by a noncommutative crepant resolution, connecting the geometry to the field theory."},{"cited_title":"Van den Bergh,Non-commutative crepant resolutions, inThe Legacy of Niels Henrik Abel: The Abel Bicentennial, Oslo, 2002, pp","cited_arxiv_id":null,"evidence_quote":"Shows existence of noncommutative crepant resolutions is equivalent to existence of crepant resolutions for isolated cDV singularities."},{"cited_title":"Morrison,The birational geometry of surfaces with rational double points., Mathematische Annalen271(1985) 415","cited_arxiv_id":null,"evidence_quote":"Classifies the possible dual graphs of exceptional curves for small resolutions, restricting the quiver topologies searched in Section 4."},{"cited_title":"Giacomelli,Rg flows with supersymmetry enhancement and geometric engineering,Journal of High Energy Physics2018(2018)","cited_arxiv_id":null,"evidence_quote":"Provides the mass-deformation counts f(X) recorded in Table 2, used to identify which singularities could have nontrivial exceptional curves."}],"review_version":1}