{"id":"0a1c13b2-9027-4bc3-a1af-74b635c782ff","arxiv_id":"1908.06036","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For (0,2) gauge theories with linear diagonal E-terms, the paper proposes a Weyl-orbifolded Landau-Ginzburg mirror and shows it reproduces quantum sheaf cohomology rings and A/2 correlation functions.","lead":"This paper proposes a construction of mirror partner theories for a class of (0,2) supersymmetric gauge theories with nonabelian gauge groups, which describe heterotic string compactifications. The proposal is tested by matching previously computed quantum cohomology rings and correlation functions, and could enable new computations of heterotic string instanton effects.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The equality of B/2 and A/2 correlators (eq. 3.9) rests on an unproven free-action claim for the Weyl orbifold plus hand-inserted Jacobian factors; the general formula may be off by a Weyl-stabilizer factor.","rationale":"The paper is explicitly a proposal, not a proof, and the authors say so in section 2 and section 3. The quantum-sheaf-cohomology half of the central claim is convincingly checked across a wide set of examples, including nonabelian Grassmannians and flag manifolds. The correlation-function half is the part that would have to be true for the mirror to actually compute heterotic Gromov-Witten-type data, and there the supporting evidence is thinner. The isolated-vacuum residue formula of [47] is a standard tool, and the reduction of (2.2) to the effective superpotential (3.7) is formal but reasonable. The genuinely shaky step is the orbifold treatment: the claim that the Weyl group acts freely on the non-excluded critical locus is asserted rather than proven, and the paper's own examples show that integrating out fields produces extra Jacobian factors that must be divided out manually to match A/2 results. That 'subtlety' is invoked after the fact, so it cannot serve as independent confirmation of the general formula. The reader's weakest_assumption identifies exactly this point, and my reading agrees. A single nonabelian check with a larger gauge group, done directly from (3.8) before field integration, would settle whether the missing Jacobian is a universal feature or an artifact of the lower-energy reductions. Because the paper's proposal remains useful and honestly hedged, the CONDITIONAL verdict stands unchanged.","tokens_in":23138,"tokens_out":10724,"duration_ms":102454,"concrete_test":"Compute for G(2,4) with generic diagonal b_i the two-point functions ⟨σ_1^2⟩, ⟨σ_1 σ_2⟩, ⟨σ_2^2⟩ in two ways: (i) from the A/2 localization formula of [18, eq. 3.63] for the original U(2) GLSM; (ii) from the untwisted-sector residue formula (3.8) applied directly to the original W-orbifold superpotential (8.4), i.e. before integrating out F_{N a}, with the Jacobian H_X H_Y included and no ad hoc 1/(1+k b_N) division. If the two disagree, or if a new hand-inserted factor is needed, the general correlator claim is falsified. A second, independent check: enumerate the fixed points of each g∈S_k on the non-excluded critical locus of (8.4) for U(3), N=5, generic b_i; if any exist, compute the twisted-sector contribution to (3.8) and verify it vanishes only under additional conditions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim has two parts: quantum sheaf cohomology relations (eq. 2.9) and correlation-function equality (eq. 3.9). The ring part is well supported: relations are reproduced for P^n×P^m, Hirzebruch surfaces, Grassmannians and flag manifolds, and the map from the mirror superpotential is mechanical. The correlator part is the weak load-bearing leg. Eq. (3.8)–(3.9) is obtained by applying the isolated-vacuum residue formula (3.6) from [47] to the W-orbifold LG model, then dropping all twisted sectors with the assertion (section 3.2) that the Weyl group acts freely on the non-excluded critical locus, 'as in [37]'. No general proof is given, and the paper states its arguments are 'somewhat formal'. The explicit checks all require a Jacobian factor (1/Δ_0 in §5.4, 1/(1+k b_N) in §8) to be divided out by hand after integrating out fields, a normalization that is justified only by the known A/2 answers. If the free-action claim fails for some gauge group or parameter choice—e.g. a Weyl conjugate pair of vacua with nontrivial stabilizer, or a value of b_i with 1+k b_i=0 where the excluded locus changes—the factor 1/|W| and the absence of twisted sectors cannot both be correct, and eq. (3.9) would disagree with A/2 correlators even when the chiral ring is reproduced. Since the correlator checks are limited to P^1×P^1, G(1,3)=P^2 and G(2,3)=P^2 and come from the same group, the equality (3.9) is currently a supported conjecture, not an established result.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a mirror construction for (0,2) GLSMs that are deformations of (2,2) GLSMs, assuming the gauge group is connected and the E-terms are linear and diagonal. The mirror is a Weyl-orbifolded (0,2) Landau-Ginzburg model with superpotential (2.2). The authors argue that the constraints of the mirror reduce to the quantum sheaf cohomology relations (2.9), and that B/2-twisted correlation functions reproduce A/2 correlation functions of the original GLSM (eqs. (3.8)-(3.9)). They verify the ring relations for P^n x P^m, Hirzebruch surfaces, Grassmannians and flag manifolds, and check correlation functions for P^1 x P^1, G(1,3) and G(2,3). The paper is explicit that this is a proposal rather than a theorem, stating in section 2 that no physical proof is claimed.","tokens_in":23426,"tokens_out":5755,"duration_ms":57677,"significance":"If the proposal is correct, it provides a systematic Hori-Vafa-style mirror for a large class of nonabelian (0,2) GLSMs, generalizing and simplifying the previous abelian construction of [36]. The paper has genuine strengths: the mirror map is parameter-free, the chiral ring relations follow from the mirror's own equations of motion in every worked example, the computations are explicit and reproducible, and the manuscript is unusually candid about the formal character of the arguments. The ring part of the claim is well supported. The correlation-function part is a supported conjecture rather than an established result, because it depends on an unproved free-orbit statement and on hand-inserted Jacobian normalizations.","major_comments":[{"comment":"The equality of B/2 and A/2 correlation functions rests on the assertion that the Weyl orbifold acts freely on the non-excluded critical locus, so that twisted sectors do not contribute and the 1/|W| factor in (3.8) is correct. This is cited to [37] but not proved, and in the present (0,2) setting the excluded locus is partly determined by the E_i(σ) = 0 conditions, which are a genuine deformation of the (2,2) case. The fixed-point argument of [37] therefore does not automatically carry over, and a stabilizer on the non-excluded critical locus would change (3.9) even when the chiral ring relations are reproduced. This missing check is load-bearing for the correlator half of the central claim.","section":"§3.2, eqs. (3.8)-(3.9)"},{"comment":"In every explicit correlator check, the lower-energy Landau-Ginzburg correlation functions come out multiplied by a Jacobian factor (1/Δ_0 in §5.4, 1/(1+k b_N) in §8) that must be divided out by hand to match the A/2 answers. The factors are mathematically natural consequences of integrating out constrained fields, but their numerical normalizations are justified only after comparison with the known A/2 results. As presented, the checks therefore calibrate the proposal rather than independently confirming eq. (3.9), leaving the general normalization in (3.9) without first-principles derivation.","section":"§5.4 and §8, eqs. (5.29), (8.22)"},{"comment":"The mirror formulas become singular at 1+k b_i = 0, and the excluded-locus condition changes at such parameter values. The paper correctly notes the relation to the bundle-versus-sheaf condition of [20, theorem 3.3], but it does not state explicitly whether the proposal, and in particular the correlation-function equality (3.9), is intended to hold when 1+k b_i = 0 or in a limiting sense as these values are approached. Since the proposal is stated for general linear diagonal E-terms, the domain of validity of eqs. (2.9) and (3.9) needs to be specified.","section":"§8, eqs. (8.9), (8.19)"},{"comment":"The free-action claim is supported by reference to [37], but [37] is itself a proposal for nonabelian (2,2) mirrors rather than an independent proof. A direct check on a theory beyond the current examples (for instance, an exceptional gauge group, a higher-rank Grassmannian, or an explicit analysis of Weyl stabilizers on the E-deformed excluded locus) would substantially strengthen the correlator claim. Without such a check, the factor 1/|W| and the neglect of twisted sectors in (3.9) remain assumptions whose failure would invalidate the equality even when the chiral ring relations are correctly reproduced.","section":"§2 and §3.2"}],"minor_comments":[{"comment":"The text states 'X_μν = σ_ν − σ_ν' along the critical locus; this appears to be a typo and should read 'X_μν = σ_ν − σ_μ' (or the sign convention made explicit), since the relevant equations of motion come from the terms 1 + (σ_μ − σ_ν)/X_μν in (8.4).","section":"§8, after eq. (8.5)"},{"comment":"The last line contains an unbalanced parenthesis: '− (d_ja_0 − b_0c_j' is missing the closing parenthesis before the factor 1/Δ_0.","section":"§5.3.2, eq. (5.36)"},{"comment":"The claim that the mirror does not depend on the details of the original superpotential, only on R-charges, is stated very briefly with a reference to [15]; a short explanation of why the construction leads to this independence would improve readability, since this is one of the paper's conjectural consequences.","section":"§10"}],"recommendation":"major_revision","confidential_remarks":"The ring-relations part of the manuscript is solid and the worked examples are convincing. The main risk is the correlation-function normalizations: the hand-inserted factors 1/Δ_0 and 1/(1+k b_N) are physically reasonable but are only checked against known answers. A more systematic derivation of these Jacobians, or at least an explicit statement that (3.9) is a conjecture whose normalization is fixed by the examples, would make the paper's status clearer. I would also note that G(1,3) and G(2,3) are both k=2 or k=1 examples with N=3, so the nonabelian correlator tests cover less ground than the ring checks; this reinforces the need for a general argument or a broader test."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is the first systematic (0,2) mirror ansatz for nonabelian GLSMs, and it does real work. The ring relations for P^n x P^m, Hirzebruch surfaces, Grassmannians and flag manifolds come out mechanically from the superpotential. The correlator matches in P^1 x P^1, G(1,3), and G(2,3) are genuine, and the paper is honest that the general argument is formal.\n\nWhat's new is the F_i(E_i(sigma) - exp(-Y_i)) terms in (2.2), which feed linear diagonal E-terms into the W-orbifold LG model. It reduces to Hori-Vafa and the authors' previous abelian and (2,2) nonabelian proposals, and it simplifies the abelian ansatz. The worked examples are substantial, including the Jacobian bookkeeping when integrating out fields; the 1/Delta_0 and 1/(1 + k b_N) factors are not hidden, they are tracked and explained.\n\nThe soft spots are in the correlator half of the claim. Equation (3.9) follows from the isolated-vacuum residue formula plus the assertion that the Weyl group acts freely on the non-excluded critical locus, so twisted sectors do not contribute. That claim is checked only in low-rank examples, and the Jacobian factors are normalized to agree with known A/2 answers. A Weyl-stabilizer or excluded-locus edge case could shift the overall normalization. The paper says it does not claim a proof, so this is a limitation of a proposal, not a hidden flaw. The chiral ring part is on much firmer ground, since the relations follow from the constraints with no adjustable parameters.\n\nThe scope is restricted to linear diagonal E-terms, connected gauge groups, and no superpotential, though section 10 sketches the superpotential case. Many tests come from the same group, so independent confirmation would help, but that is not a defect by itself.\n\nBottom line: this deserves a serious referee. It advances the field and the checks are meaningful. I would recommend publishing after the correlator justification is either strengthened or explicitly flagged as conjectural in general. For a reading group, it is worth a session; I would cite it if I worked on (0,2) mirrors.","headline":"First systematic nonabelian (0,2) mirror proposal; ring checks are solid, correlator checks are convincing but leave a free-action assumption to verify.","tokens_in":24127,"tokens_out":1992,"would_cite":true,"duration_ms":19377,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proposes a Weyl-orbifolded Landau-Ginzburg mirror for (0,2) GLSMs with linear diagonal E-terms, and argues, with example checks, that its constraints reproduce quantum sheaf cohomology while its B/2 correlation functions match…","keywords":["(0,2) mirror symmetry","gauged linear sigma model","quantum sheaf cohomology","Landau-Ginzburg mirror","Weyl orbifold","A/2 twist","B/2 twist","nonabelian mirror"],"falsifier":"Compute the B/2 correlation functions of the proposed mirror for a Grassmannian $G(3,4)$ with generic diagonal deformations $b_i$, compare them with the A/2 localization results, and check whether every correlator matches after multiplying the naive Landau-Ginzburg result by exactly $1/(1+3b_4)$; a single mismatch, or a failure of the quantum sheaf cohomology relations at a point where the orbifold symmetry has a fixed point on the non-excluded locus, would falsify the proposal.","tokens_in":22784,"feed_emoji":"🔁","tokens_out":16242,"duration_ms":138259,"temperature":0.7,"pith_summary":"This paper proposes a mirror construction for two-dimensional (0,2) supersymmetric gauge theories (GLSMs) that are deformations of ordinary (2,2) theories. The mirror is a Weyl-group orbifold of a Landau-Ginzburg model whose superpotential is built from the weights of the matter representation, the roots of the gauge group, and the linear diagonal E-terms $D_+\\Psi_i=E_i(\\sigma)\\Phi_i$. The paper's central claim is that the mirror's constraint equations reproduce the quantum sheaf cohomology relations of the original theory, the heterotic analogue of quantum cohomology, and that its B/2-twisted correlation functions match the A/2-twisted correlation functions in every example checked. The authors state explicitly that this is a proposal with consistency tests, not a proof. If correct, it gives a direct route from a (0,2) gauge theory to a Landau-Ginzburg model in which worldsheet instanton corrections become ordinary residue computations.","feed_headline":"Mirror superpotential reproduces (0,2) quantum sheaf cohomology","feed_subtitle":"A Weyl-orbifold Landau-Ginzburg model matches A/2 correlation functions in every tested abelian and nonabelian example.","key_machinery":"The central object is the Weyl-orbifolded Landau-Ginzburg superpotential (2.2), built from pairs $(Y_i,F_i)$, $(\\sigma_a,\\Upsilon_a)$, and $(X_{\\tilde\\mu},\\Lambda_{\\tilde\\mu})$, with the terms $Y_i$ coupled to weights $\\rho^a_i$, $X_{\\tilde\\mu}$ to roots $\\alpha^a_{\\tilde\\mu}$, and $F_i$ enforcing $e^{-Y_i}=E_i(\\sigma)$. The orbifold group simultaneously permutes and signs the members of each pair, and the proposal claims it acts freely on the non-excluded critical locus. What makes this machinery work is that integrating out $X_{\\tilde\\mu}$ and $Y_i$ produces an effective superpotential whose critical-point equations are exactly the quantum sheaf cohomology relations, while the determinant factors $H_X$ and $H_Y$ from the integration feed into the residue formula $\\langle f\\rangle=\\sum f/\\det(\\partial_i J_j)$, which reproduces the A/2 correlation function formula of the original GLSM.","core_discovery":"For a (0,2) GLSM with connected gauge group $G$ of dimension $n$ and rank $r$, matter in a representation $R$, and E-terms of the linear diagonal form $D_+\\Psi_i=E_i(\\sigma)\\Phi_i$, the paper proposes that the mirror is the Weyl-group orbifold of a (0,2) Landau-Ginzburg model with superpotential (2.2). The Fermi-field constraints give the operator mirror map $e^{-Y_i}=E_i(\\sigma)$ and $X_{\\tilde\\mu}=\\sum_a \\alpha^a_{\\tilde\\mu}\\sigma_a$; exponentiating the D-term constraints turns these into $\\prod_i E_i(\\sigma)^{\\rho^a_i}=\\tilde q_a$, which are precisely the quantum sheaf cohomology relations of the original theory. The B/2-twisted correlation functions computed from the mirror through the isolated-vacuum residue formula are argued to equal the A/2 correlation functions of the original GLSM, and the match is verified in detail for $P^n\\times P^m$, Hirzebruch surfaces, and the first Grassmannian cases $G(1,3)$ and $G(2,3)$. In the nonabelian checks a Jacobian factor must be divided out by hand to get the agreement. The paper calls the construction a proposal, not a theorem, and gives only consistency tests rather than a general proof.","pith_inferences":["If the proposal survives tests beyond the cases checked here, it would turn heterotic worldsheet instanton sums into residue evaluations in a Landau-Ginzburg model, a concrete program the paper motivates but does not carry out.","The diagonal restriction is probably not a hard limit: if off-diagonal linear E-terms are irrelevant for A/2-twisted correlators, a basis change could bring any linear E-term into diagonal form; the paper only states the ansatz for the diagonal case.","The hand-divided Jacobian factors suggest an unstated normalization convention in the path-integral derivation; finding a systematic rule for these factors would likely upgrade the proposal from example-by-example matching to a proof.","Testing the same Weyl-orbifold machinery on O(k) or exceptional gauge groups, where the paper notes the extension is straightforward but does not carry it out, would probe the free-action assumption more sharply than the Grassmannian cases."],"forward_implications":["On the (2,2) locus the ansatz reduces to the standard abelian mirror prescription and to the nonabelian mirror proposal it extends, so every linear diagonal (0,2) deformation inherits a Landau-Ginzburg mirror of the same general form.","For abelian theories the proposal simplifies and generalizes the earlier systematic mirror construction, and it reproduces the earlier superpotentials when restricted to the same class of deformations.","Quantum sheaf cohomology relations are reproduced for $P^n\\times P^m$, Hirzebruch surfaces, Grassmannians $G(k,N)$, and flag manifolds, including the phase shifts generated by integrating out the $X$ fields.","In every tested example the B/2 correlation functions agree with the A/2 correlation functions after dividing out the Jacobian factor the paper computes explicitly.","For theories with a (0,2) superpotential the mirror depends only on R-charges, not on the detailed superpotential, consistent with the claim that A/2-twisted GLSMs are insensitive to such details."],"supporting_citations":[{"why":"supplies the (2,2) nonabelian mirror construction, including the Weyl orbifold action and the $X$ fields, that this paper extends to (0,2) theories.","marker":"[37]"},{"why":"supplies the isolated-vacuum correlation formula $\\langle f\\rangle = \\sum f/\\det(\\partial_i J_j)$ used for the B/2 Landau-Ginzburg correlators.","marker":"[47]"},{"why":"supplies the A/2 correlation function formula and quantum sheaf cohomology relations that the mirror correlators are claimed to match.","marker":"[18]"},{"why":"supplies the previous abelian (0,2) mirror ansatz that the current proposal generalizes and simplifies.","marker":"[36]"},{"why":"supplies the $P^n\\times P^m$ Toda-like (0,2) mirrors and A/2 correlation functions used as a comparison in section 5.","marker":"[34]"},{"why":"supplies the Grassmannian quantum sheaf cohomology relations and A/2 correlation functions used to verify the nonabelian ansatz.","marker":"[19]"},{"why":"supplies the bundle-versus-sheaf condition $1+kb_i\\neq 0$ that shapes the Grassmannian mirror's excluded locus.","marker":"[20]"},{"why":"supplies the flag manifold quantum sheaf cohomology relations reproduced in section 9.","marker":"[21]"},{"why":"supplies the Coulomb-branch effective superpotential and the R-charge dependence used for hypersurface mirrors.","marker":"[15]"},{"why":"supplies the toric quantum sheaf cohomology relations used for $P^n\\times P^m$ and Hirzebruch surfaces.","marker":"[17]"}],"fun_headline_variants":["Weyl-orbifold mirrors match (0,2) GLSM correlation functions","Nonabelian (0,2) mirrors via Weyl-orbifold LG pass tests","Mirror proposal reproduces (0,2) quantum sheaf cohomology in examples","Weyl-orbifold LG mirrors for (0,2) GLSMs: checks in Grassmannians","Proposal: (0,2) mirrors via Weyl-orbifold LG match quantum cohomology"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The correlation-function half of the proposal rests on the assumption that the mirror's orbifold symmetry acts freely on the non-excluded vacuum solutions, so that no extra sectors contribute, and on the expectation that the normalization mismatch seen in every example is cured by dividing out a single Jacobian factor.","fun_headline_variants_meta":{"raw":{"variants":["Weyl-orbifold mirrors match (0,2) GLSM correlation functions","Nonabelian (0,2) mirrors via Weyl-orbifold LG pass tests","Mirror proposal reproduces (0,2) quantum sheaf cohomology in examples","Weyl-orbifold LG mirrors for (0,2) GLSMs: checks in Grassmannians","Proposal: (0,2) mirrors via Weyl-orbifold LG match quantum cohomology"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001388,"raw_usage":{"total_tokens":5635,"prompt_tokens":977,"completion_tokens":4658,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":593,"completion_tokens_details":{"reasoning_tokens":4530}},"tokens_in":593,"tokens_out":4658,"duration_ms":28387,"temperature":1.0,"reasoning_tokens":4530,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:59:04.276579+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the B/2 correlation functions of the proposed mirror for a Grassmannian $G(3,4)$ with generic diagonal deformations $b_i$, compare them with the A/2 localization results, and check whether every correlator matches after multiplying the naive Landau-Ginzburg result by exactly $1/(1+3b_4)$; a single mismatch, or a failure of the quantum sheaf cohomology relations at a point where the orbifold symmetry has a fixed point on the non-excluded locus, would falsify the proposal.","supporting_citations":[{"cited_title":"Half-Twisted (0,2) Landau-Ginzburg Models","cited_arxiv_id":"0712.1058","evidence_quote":"supplies the isolated-vacuum correlation formula $\\langle f\\rangle = \\sum f/\\det(\\partial_i J_j)$ used for the B/2 Landau-Ginzburg correlators."},{"cited_title":"A proposal for (0,2) mirrors of toric varieties","cited_arxiv_id":"1707.05274","evidence_quote":"supplies the previous abelian (0,2) mirror ansatz that the current proposal generalizes and simplifies."},{"cited_title":"Toda-like (0,2) mirrors to products of projective spaces","cited_arxiv_id":"1603.09634","evidence_quote":"supplies the $P^n\\times P^m$ Toda-like (0,2) mirrors and A/2 correlation functions used as a comparison in section 5."},{"cited_title":"Quantum sheaf cohomology on Grassmannians","cited_arxiv_id":"1512.08586","evidence_quote":"supplies the Grassmannian quantum sheaf cohomology relations and A/2 correlation functions used to verify the nonabelian ansatz."},{"cited_title":"Classical sheaf cohomology rings on Grassmannians","cited_arxiv_id":"1605.01410","evidence_quote":"supplies the bundle-versus-sheaf condition $1+kb_i\\neq 0$ that shapes the Grassmannian mirror's excluded locus."},{"cited_title":"Summing the Instantons in Half-Twisted Linear Sigma Models","cited_arxiv_id":"0810.0012","evidence_quote":"supplies the Coulomb-branch effective superpotential and the R-charge dependence used for hypersurface mirrors."},{"cited_title":"Physical aspects of quantum sheaf cohomology for deformations of tangent bundles of toric varieties","cited_arxiv_id":"1110.3752","evidence_quote":"supplies the toric quantum sheaf cohomology relations used for $P^n\\times P^m$ and Hirzebruch surfaces."}],"review_version":1}