{"id":"7e175060-1cc7-4ef8-b562-81ab1b7076e1","arxiv_id":"2502.08215","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Massive IIA flux compactifications on singular and Joyce G2 orbifolds yield scale-separated AdS3 vacua whose dual operator dimensions are integer up to parametrically small corrections.","lead":"This paper builds new three-dimensional string vacua in which the extra dimensions are much smaller than the spacetime curvature radius, and finds that the associated dual field theory operators can have whole-number scaling dimensions. The construction uses a special seven-dimensional Joyce orbifold, whose known smoothing may make the vacuum easier to trust, and is relevant to the swampland debate about which low-energy theories can come from quantum gravity.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Internally consistent, but the physical vacuum is not established: the smeared intersecting-O6 background may not lift to a localized solution.","rationale":"I checked the core algebra and found it internally consistent: the superpotential (3.10), the supersymmetric solution (3.11), and the normalized Hessian (3.16) together give the eigenvalues that produce Delta = {8, 4, 4, 4, 4, 2, 2, 2} in eq. (3.17). The matrix displayed in eq. (4.18) is V_AB/|V|, and composing it with the kinetic matrix (4.19) gives trace(K^{-1}V) = 80, matching the claimed eigenvalue sum {48, 8, 8, 8, 8, 0, 0, 0}; so I do not see an internal algebraic failure. The central risk is physical: the construction uses smeared, intersecting O6-planes and only the untwisted closed-string sector, and both limitations are acknowledged in the paper. Since reference [23] indicates that localized intersecting sources can fail to solve the supergravity equations, the existence of the vacuum is the decisive open question. This is exactly the weakest assumption identified by the reader, and my stress-test does not introduce a new objection or overturn the CONDITIONAL verdict: the algebraic derivation supports the formal claim, but the claim about actual string vacua should remain conditional pending evidence that the smeared limit is reliable or that a blow-up removes the O6 intersections.","tokens_in":15448,"tokens_out":17617,"duration_ms":191378,"concrete_test":"Compute the first-order localized backreaction for the Joyce orbifold (c = 1/2) with the O2 plus the three O6-planes of eq. (2.7) and fluxes H3 = h(Phi1 + Phi2 + Phi3), F0 = m, F4 = -f(Psi4 + Psi5 + Psi6 + Psi7), extending the AdS3 O6 backreaction analysis of [44] to intersecting O6s. Check whether a perturbative localized solution exists for parametrically large f and whether the corrected mass eigenvalues of the eight untwisted scalars differ from {48, 8, 8, 8, 8, 0, 0, 0} in units of |V|; if the localized solution fails to exist or the O(1) eigenvalues shift, the integer spectrum is an artifact of smearing rather than a property of a genuine vacuum.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The integer conformal dimensions are derived inside the classical smeared-orientifold truncation. The load-bearing condition is not an algebraic step in the Hessian computation but the existence of the 10D background: if the mutually intersecting O6-planes do not admit a localized supergravity solution in the limit that should recover the smeared vacuum, the computed spectrum is the spectrum of an effective theory that does not describe a string vacuum. The paper itself flags this in Sections 2.1 and 5, and reference [23] reports that localized intersecting sources can fail to admit supergravity solutions. Reference [24] suggests that blowing up the orbifold singularities might remove the intersections, but the blow-up is not carried out here, and the present analysis remains in the untwisted sector. The analytic full-moduli-stabilization solution of Section 4 stabilizes moduli within the same smeared approximation and therefore does not resolve the backreaction question. Thus the central claim, as a statement about actual AdS3 string vacua, rests on an unresolved physical assumption rather than on the internal consistency of the effective theory, which appears sound.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies supersymmetric, scale-separated AdS3 flux vacua of massive IIA on T7/Z2^3 G2 orbifolds with smeared O2/O6 planes, comparing a 'singular' orbifold with c=0 and a Joyce-type orbifold with c=1/2. Using the 3D N=1 superpotential of [43], the authors specialize to H3 parallel to the O6 currents, Romans mass F0, and specific F4 components, solve the F-term equations analytically, and verify weak coupling, large internal volume, and parametric scale separation. The normalized Hessian gives m^2 L^2 = {48,8,8,8,8,0,0,0}, hence conformal dimensions Delta = {8,4,4,4,4,2,2,2} in the classical smeared untwisted sector. An anisotropic F4 variant is shown to preserve these dimensions. In Section 4, the authors add extra F4 components and adjust the D2-brane/tadpole contribution, breaking the flat direction of the simpler solution; with N = (2/5)G^2 and n = 5/G they find an analytic one-parameter family that stabilizes the eight scalars of the truncation. As G tends to infinity, the kinetic and mass matrices approach those of the integer-dimensional solution, so the dual dimensions are parametrically close to integers while g_s tends to zero and the volume diverges. The paper explicitly states that the backreaction of localized intersecting O6-planes and the twisted sector are not addressed.","tokens_in":15517,"tokens_out":12791,"duration_ms":141382,"significance":"The algebraic core of the paper is coherent: no free parameter is fitted to obtain the integer spectrum, the flux quantization conditions are made explicit, the Section 4 solution is fully analytic, and the parametric scalings are consistent with the stated limits. If the smeared orientifold approximation is a reliable leading-order description, or if the Joyce blow-up removes the O6 intersections as suggested by [24], the paper would be a distinctive addition to the AdS3 scale-separation literature, extending the known integer-conformal-dimension pattern to minimal-supersymmetry AdS3 vacua. The paper is also commendably transparent about its main weaknesses, namely the unresolved backreaction of intersecting smeared O6-planes and the restriction to the untwisted sector. The significance is therefore conditional on a physical assumption that is not established in the manuscript.","major_comments":[{"comment":"The load-bearing physical assumption is the existence of the localized 10D background. Equations (2.8)-(2.10) impose tadpole cancellation with smeared source currents, and the entire mass spectrum is computed in the effective theory built from this smeared configuration. Sections 2.1 and 5 acknowledge that the backreaction of the mutually intersecting O6-planes remains open, and reference [23] shows that localized intersecting sources can fail to admit a supergravity solution. Since the title and abstract present these as scale-separated AdS3 vacua of massive IIA, the manuscript should either provide evidence that the smeared limit is a controlled leading-order approximation, for example through a first-order backreaction analysis along the lines of [44] or through an explicit blow-up of the Joyce orbifold showing that the sources cease to intersect, or it should rescope the conclusions to the smeared classical truncation and state clearly that the string-vacuum realization is an open assumption. As it stands, the existence of the vacuum is an assumption rather than a result.","section":"Sec. 2.1, eqs. (2.8)-(2.10); Sec. 5"},{"comment":"The exact integer values are recovered only in the limit G -> infinity, and this is the same limit in which the determinant in (4.14) vanishes; the fully stabilized solution therefore degenerates at the point where the spectrum is exactly integer. For every finite G the stabilized vacuum has dimensions that are only parametrically close to integers. The paper is transparent about this, but the logical structure should be made explicit in the abstract and introduction: the exactly integer spectrum (3.17) is a property of the simpler, classically unstabilized vacuum of Section 3, while the fully stabilized vacuum of Section 4 has approximate integers and tends to that unstabilized solution only as G -> infinity. As written, the reader may infer that a single fully stabilized AdS3 vacuum has exactly integer conformal dimensions.","section":"Sec. 4.2, eqs. (4.14)-(4.20)"}],"minor_comments":[{"comment":"The first sentence says the flux choices 'yield integer dual dimensions,' while the last sentence says the dimensions are 'only parametrically close to integer values.' This is consistent once the formal limit is specified, but the phrasing invites confusion; the abstract should state explicitly that exact integers occur only in the classical smeared truncation or in the formal G -> infinity limit.","section":"Abstract"},{"comment":"The object displayed in (4.18) is called the 'normalized Hessian,' but it appears to be the second-derivative matrix divided by |V|; the physical mass eigenvalues require multiplication by K^{-1} as in (3.16). Since the kinetic matrix is given separately in (4.19), the text should clarify that the physical mass matrix is K^{-1} V / |V| in the G -> infinity limit.","section":"Sec. 4.2, eq. (4.18)"},{"comment":"The radii r_i in (3.24) are introduced without a definition; the text should state whether they are string-frame or Einstein-frame radii and to which torus coordinates they refer.","section":"Sec. 3.3, eq. (3.24)"},{"comment":"Reference [29] is incomplete, as it lacks a title. In addition, the symbol N is used both for the flux quantum in (2.15) and for the large F4 flux in Section 4; please disambiguate these two uses.","section":"References and notation"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's main risk is the well-known smeared-orientifold caveat, which the authors do not hide. If the journal requires an actual localized 10D solution before publication, this paper would need substantial additional work; otherwise, it is a reasonable formal contribution to the scale-separation program. The self-citation to [43] for the superpotential is legitimate but should be kept in mind when assessing the novelty of the derivation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is a solid effective-field-theory construction of scale-separated AdS3 vacua with integer dual dimensions in the formal limit, and it improves on the earlier G2-orientifold work by finding flux choices that actually give the integer spectrum and by adding a Joyce orbifold whose known resolution might sidestep intersecting-O6 issues. The algebra I checked is coherent; the caveats are the ones the authors themselves flag.\n\nWhat's new: the specific flux choices producing Δ={8,4,4,4,4,2,2,2} in both the singular and Joyce T^7/Z2^3 orbifolds, and the fully stabilized analytic solution parameterized by a single integer G, which approaches the integer-dimension spectrum as G→∞. The superpotential is imported from [43], but the extremization and mass spectrum are derived here, and the integer dimensions are an output, not an input. The Joyce orbifold observation is genuinely interesting: because the involutions other than Θα, Θβ, Θγ act freely for c=1/2, the singular set is smaller, and after blow-up the O6-planes may no longer intersect. That is a concrete way to address the main consistency worry.\n\nSoft spots, in order of importance. First, the construction lives in the smeared-orientifold truncation. The mass spectrum, the moduli stabilization, and the integer dimensions are all computed in that effective theory. Whether a fully localized solution exists with mutually intersecting O6-planes is not settled; [23] gives reason to worry, and blow-up is not carried out here. The authors say this openly in Sections 2.1 and 5, so it is a limitation, not a hidden flaw. Second, the integer dimensions are formal-limit statements; higher-order corrections shift them, which is standard for this literature and acknowledged. Third, the Hessian and kinetic matrices at large G are presented as leading-order results with O(1/G) corrections not shown in detail. That is fine for the claim, but it makes verification slower.\n\nThe citation pattern is honest; self-citation to [43] is appropriate since the superpotential comes from there. I found no circularity. The paper is for people working on scale separation and swampland; it extends the census of AdS3 examples with integer dual dimensions. It deserves a serious referee.","headline":"Solid EFT-level construction with genuinely new flux choices and a useful Joyce-orbifold idea; the smeared-backreaction caveat is real but openly acknowledged.","tokens_in":16141,"tokens_out":2069,"would_cite":true,"duration_ms":19779,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"In the formal limit, all eight closed-string scalar dual dimensions are integers {8,4,4,4,4,2,2,2}.","keywords":["AdS3 flux vacua","massive type IIA","G2 orbifolds","scale separation","integer conformal dimensions","moduli stabilization","smeared orientifolds","T^7/Z2^3 orbifolds"],"falsifier":"Compute the first-order backreaction of the localized O6-planes for these flux choices---for example, by finding localized profiles in the singular and desingularizable orbifolds---and recompute the normalized Hessian of the scalar potential. If the mass eigenvalues deviate from $\\{48,8,8,8,8,0,0,0\\}$ by a nonvanishing amount that does not vanish as $G\\to\\infty$, the integer spectrum is an artifact of the smeared approximation and fails as a statement about the actual vacuum.","tokens_in":15137,"feed_emoji":"🔢","tokens_out":15601,"duration_ms":138139,"temperature":0.7,"pith_summary":"This paper studies three-dimensional anti-de Sitter (AdS$_3$) flux vacua of massive type IIA string theory (the ten-dimensional supergravity with a nonzero $F_0$ flux) on G2 orbifolds, and asks whether the light closed-string scalar fields have operator dimensions in the conjectured dual CFT that are integers, a pattern seen in other scale-separated string vacua. The authors show that on two closely related $T^7/\\mathbb{Z}_2^3$ orbifolds with the orientifold planes smeared rather than localized, specific flux choices make every untwisted scalar satisfy $\\Delta_A = 1+\\sqrt{1+m_A^2 L^2} = \\{8,4,4,4,4,2,2,2\\}$ in the formal limit where the string coupling vanishes and the internal volume is infinite. One of the two orbifolds admits a known desingularization into a smooth G2 manifold, and resolving the singularities may remove the intersecting orientifold planes that otherwise threaten the smeared-source approximation. The paper also exhibits a flux configuration with all moduli stabilized; its mass spectrum approaches these integers parametrically as a large integer $G$ tends to infinity, so the integer dimensions are statements about a limit rather than about finite flux values.","feed_headline":"String vacuum yields integer conformal dimensions","feed_subtitle":"Two G2-orbifold flux choices give AdS3 scalars dimensions 8,4,4,4,4,2,2,2, extending a pattern seen across flux vacua.","key_machinery":"The load-bearing object is the 3D $\\mathcal{N}=1$ effective superpotential for G2 orientifold compactifications, $$P = \\frac{m}{8}$e^{{\\frac{y}}${2}-\\frac{\\sqrt{7}}{2}x}+\\frac{h}{8}$e^{{y+\\frac{x}}${\\sqrt{7}}}\\left(\\frac{1}{\\tilde{s}_1}+\\frac{1}{\\tilde{s}_2}+\\frac{1}{\\tilde{s}_3}\\right)-\\frac{f}{8}$e^{{y-\\frac{x}}${\\sqrt{7}}}\\left(\\tilde{s}_4+\\tilde{s}_5+\\tilde{s}_6+\\frac{1}{\\prod_{a=1}^6\\tilde{s}_a}\\right),$$ together with the harmonic 3-form and 4-form bases $\\Phi_i,\\Psi_i$ of the G2 orbifold and a shift parameter $c\\in\\{0,1/2\\}$ distinguishing the two orbifold types. Extremizing $P$ fixes the eight scalar fields $\\phi^A=(x,y,\\tilde{s}_1,\\ldots,\\tilde{s}_6)$; the masses are the eigenvalues of the normalized Hessian $\\langle K^{AB}\\rangle^{-1}\\langle V_{BC}\\rangle/|\\langle V\\rangle|$, and the relation $\\Delta_A = 1+\\sqrt{1+m_A^2 L^2}$ converts those eigenvalues into dual conformal dimensions. The second key ingredient is a generalized $F_4$ flux with an integer parameter $G$ that breaks a shift symmetry, removes the flat direction, and reduces to the simpler integer-dimension flux as $G\\to\\infty$.","core_discovery":"The central claim is that the untwisted closed-string scalar sector of these scale-separated AdS$_3$ vacua can be arranged by choosing fluxes so that every scalar mass squared in units of the AdS scale matches the values $$$m_A^{2}$ $L^{2}$ = \\{48,8,8,8,8,0,0,0\\},$$ which through the AdS/CFT relation $\\Delta_A = 1+\\sqrt{1+m_A^2 L^2}$ gives the integer conformal dimensions $$\\Delta_A = \\{8,4,4,4,4,2,2,2\\}.$$ This is shown for both the singular $c=0$ orbifold and the desingularizable $c=1/2$ orbifold, which admit the same flux ansatz and produce the same untwisted spectrum. The paper further shows that adding $F_4$ flux along the remaining harmonic four-forms lifts the flat direction of the simpler flux choice and stabilizes all eight moduli, with the normalized Hessian converging to the matrix that yields the integer spectrum as a single large integer flux $G\\to\\infty$; at any finite $G$ the dimensions are only parametrically close to integers, exactly as in all other known examples.","pith_inferences":["Beyond the paper: if the integer spectrum survives in the resolved orbifold, a first-order backreaction calculation in this construction would provide a concrete test of whether the smeared approximation is the leading-order description of a real vacuum.","Beyond the paper: the same $\\Delta_A$ values appearing for both orbifolds and being independent of the anisotropy parameter $n$ suggests the integers are fixed by the G2 structure-form topology and the tadpole algebra, not by flux magnitudes; enumerating all flux vectors satisfying the same Bianchi identities could test this.","Beyond the paper: the pattern $\\{8,4,4,4,4,2,2,2\\}$ may reflect the arithmetic of the $\\mathbb{Z}_2^3$ action on the seven-torus, with the $c=1/2$ shift making some involutions freely acting; a scan over $\\mathbb{Z}_2^k$ orbifolds could reveal whether integer dimensions always accompany such shifts.","Beyond the paper: since only the untwisted sector is treated, computing twisted-sector masses after blowing up the singularities would show whether integer dimensions are a property of the putative dual CFT or an artifact of the truncation; the paper leaves this open."],"forward_implications":["In the formal limit $G\\to\\infty$, the fully stabilized vacuum has $g_s\\sim G^{-3/2}$, string-frame volume $\\sim G^{7/2}$, and $L_{\\mathrm{KK}}^2/L_{\\mathrm{AdS}}^2\\sim G^{-2}$, so weak coupling, large volume, and parametric scale separation coexist.","The singular $c=0$ and desingularizable $c=1/2$ orbifolds give the same untwisted closed-string spectrum and the same integer dual dimensions for the same flux choices.","The normalized mass eigenvalues $\\{48,8,8,8,8,0,0,0\\}$ translate to dual dimensions $\\{8,4,4,4,4,2,2,2\\}$; this equality is exact only in the asymptotic limit, with corrections of order $1/G$ at finite flux.","The simple flux choice of Section 3 leaves one flat direction that would be lifted by higher-order corrections; the more general flux of Section 4 stabilizes all eight moduli while preserving the asymptotic integer spectrum.","If the known desingularization of the $c=1/2$ orbifold removes orientifold intersections, the smeared-source approximation may be more defensible in this construction than in the singular case."],"supporting_citations":[{"why":"Constructs the original scale-separated AdS3 flux vacua on G2 orientifolds and supplies the 3D effective superpotential this paper starts from.","marker":"[43]"},{"why":"Provides the known desingularization of the c=1/2 orbifold into smooth compact G2 spaces, which distinguishes the two orbifold types.","marker":"[51]"},{"why":"Shows that orientifold O6 intersections can disappear after blowing up orbifold singularities, the basis for the expectation that the desingularizable orbifold may avoid intersecting sources.","marker":"[24]"},{"why":"Establishes that higher-order corrections shift masses and spoil integer conformal dimensions, fixing the asymptotic sense in which the present result holds.","marker":"[39]"},{"why":"Computes first-order O6-plane backreaction on the AdS3 vacua, the relevant check of the smeared-source approximation used here.","marker":"[44]"},{"why":"Documents integer conformal dimensions in type IIA flux vacua, the pattern the paper extends to these AdS3 constructions.","marker":"[35]"},{"why":"Notes that dual operator dimensions remain unchanged under anisotropic flux choices, supporting the n-independence found in Section 3.3.","marker":"[49]"}],"fun_headline_variants":["AdS3 flux vacua give integer conformal dimensions","G2 orbifold flux choices force integer scalar dimensions","Scale-separated AdS3: integer dimensions from massive IIA","AdS3 scalars: 8,4,4,4,4,2,2,2 from flux"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The computation treats the orientifold planes as spread out uniformly rather than concentrated along their worldvolumes; if their true localized backreaction cannot be captured this way, or if it shifts the untwisted scalar masses, the integer conformal dimensions would not describe a real string vacuum.","fun_headline_variants_meta":{"raw":{"variants":["AdS3 flux vacua give integer conformal dimensions","G2 orbifold flux choices force integer scalar dimensions","Scale-separated AdS3: integer dimensions from massive IIA","AdS3 scalars: 8,4,4,4,4,2,2,2 from flux"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000742,"raw_usage":{"total_tokens":3287,"prompt_tokens":896,"completion_tokens":2391,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":512,"completion_tokens_details":{"reasoning_tokens":2308}},"tokens_in":512,"tokens_out":2391,"duration_ms":17608,"temperature":1.0,"reasoning_tokens":2308,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T06:00:42.211110+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the first-order backreaction of the localized O6-planes for these flux choices---for example, by finding localized profiles in the singular and desingularizable orbifolds---and recompute the normalized Hessian of the scalar potential. If the mass eigenvalues deviate from $\\{48,8,8,8,8,0,0,0\\}$ by a nonvanishing amount that does not vanish as $G\\to\\infty$, the integer spectrum is an artifact of the smeared approximation and fails as a statement about the actual vacuum.","supporting_citations":[{"cited_title":"Constructing compact manifolds with except ional holonomy,","cited_arxiv_id":null,"evidence_quote":"Provides the known desingularization of the c=1/2 orbifold into smooth compact G2 spaces, which distinguishes the two orbifold types."}],"review_version":1}