{"id":"05716a6d-276a-42e0-809c-51507c1b0234","arxiv_id":"2509.09039","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Boundary W-algebras W(sl_n[um,s], n/u) and principal W-algebras W(sl_s[s], s/u) have matching q-characters and modular data, hence identical fusion rules.","lead":"This paper proves that certain pairs of boundary W-algebras in type A have identical lists of irreducible modules, identical characters, and identical modular data, forcing the same fusion rules. The result determines fusion rules for a broad family of exceptional W-algebras and reduces the whole problem to the classical principal case.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proposition 4.20's reduction to a W_f sum may rely on an unjustified orthogonality step; this is load-bearing for the S-matrix equality and fusion rules.","rationale":"The reader's weakest assumption concerns uniqueness of the minimal conformal dimension. This is not load-bearing: Definition 2.9 already requires an index ◦ with S_{◦,i} > 0 for any modular datum, and Huang's theorem gives the W-algebra category an MTC structure, so positivity is automatic. The comparison in Lemma 2.11 uses only that axiom, not conformal dimension. The claim that Section 4.3.2 verifies uniqueness is inaccurate but harmless.\n\nA more serious issue is the derivation of formula (4.4) in Proposition 4.20. The step replacing w_f w_0 β' by w_f β' in the exponent is not justified by the cited orthogonality with Δ_0. In the pyramid, W_0 (column symmetries) and W_f (row symmetries) do not commute, and the orthogonal complement of Δ_0 is not W_f-invariant. Without (4.4), the subsequent factorization and comparison to the principal W-algebra S-matrix do not go through. This directly affects Theorem 1.1(3) and Theorem 4.12. The result may be true and the step salvageable with additional work, but as written the proof is incomplete. Therefore I recommend keeping the verdict conditional, though for a different reason than the reader's.","tokens_in":28540,"tokens_out":25733,"duration_ms":240420,"concrete_test":"For the smallest nontrivial case n=11, u=8, s=3, compute the S-matrix entries S_{λ,λ'} both by the original formula (3.4)/(4.3) with explicit choices of y making λ g_0-integrable, and by the reduced formula (4.4). If the values differ, Proposition 4.20 is false. Alternatively, directly test the identity (β, w_f w_0 β') = (β, w_f β') for the contributing w_f ∈ W_f and w_0 ∈ W_0 in that example; if it fails, the proof has a concrete counterexample.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Proposition 4.20 reducing (4.3) to (4.4) contains an unjustified step. After Corollary 4.19, the sum is over w_f w_0 ∈ W_f W_0. The exponent is (β, w_f w_0 β'), and the paper replaces it by (β, w_f β'), citing (β, Δ_{0,+}) = 0 to see that w_0(β) = β. But w_0 does not act on β in the exponent; the correct manipulation is (β, w_f w_0 β') = (w_f^{-1}β, w_0 β'). Since w_0 β' - β' lies in the span of Δ_0, the replacement requires (w_f^{-1}β, α) = 0 for α ∈ Δ_0. No such orthogonality is proved; W_f (row permutations) does not in general preserve the orthogonal complement of Δ_0 in the pyramid geometry, as can be seen already for n=11, u=8, s=3. Thus formula (4.4) is not derived. Because (4.4) is the pivotal identity that later yields the comparison with the principal W(sl_s[s], s/u) S-matrix, the equality of S-matrices in Theorem 4.12 and hence the fusion rules are not established as written.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies exceptional (boundary) W-algebras in type A, focusing on the family W(sl_n[u^m,s], n/u) with n=um+s, 1≤s≤u−1, gcd(s,u)=1. Its main theorem (Theorem 1.1 and Theorem 4.12) asserts that the irreducible modules of this boundary W-algebra are in bijection with those of the principal W-algebra W(sl_s[s], s/u), that the q-characters agree under the bijection, and that the modular data, hence the fusion rules, coincide for all m. The proof combines the Kac–Wakimoto character formula (Proposition 3.1), the parametrization of admissible weights by necklaces (Theorem 4.6), equality of q-characters via pyramid combinatorics (Theorem 4.8), and a Weyl-group manipulation of the S-matrix (Propositions 4.16 and 4.20) together with conformal-dimension computations (Section 4.3.2). As an application, the paper derives a factorization of fusion rules for general exceptional W-algebras in type A (Corollaries 1.2 and 4.14), gives a conceptual explanation of product formulas for certain Virasoro characters in type D (Proposition 3.2), and presents numerical modular data for the type E_8 subregular case W(E_8(a_1),31/29).","tokens_in":28814,"tokens_out":23046,"duration_ms":243535,"significance":"If the main theorem is correct, it is a substantial structural result: it reduces the modular data and fusion rules of all boundary exceptional W-algebras in type A to the well-studied principal W-algebras, and hence, together with the factorisation corollary, gives a largely complete determination of fusion rules for type A exceptional W-algebras. The paper also contains valuable technical contributions: the explicit necklace parametrization, the m-independence of q-characters, the S-matrix factorisation, and the computational treatment of the E_8 subregular example. A notable strength is that many of the intermediate claims are supported by explicit formulas and finite computations rather than by abstract existence arguments. However, two load-bearing steps in the written proof need correction: the reduction in Proposition 4.20 is misstated, and the uniqueness of the minimal conformal dimension, used to fix signs via Lemma 2.11, is not actually verified in the section cited for it.","major_comments":[{"comment":"The derivation of (4.4) from (4.3) is not correct as printed. In (4.3) the exponent is (β, w(β')), so after restricting to w = w_f w_0 it is (β, w_f w_0(β')), not ((w_f w_0)(β), β'). The proof replaces this with (w_f(β), β') and justifies it by (β, Δ_{0,+})=0, but w_0 acts on β' in the exponent, not on β. A valid repair exists: Lemma 4.4 gives (β', Δ_0)=0, hence w_0(β')=β', so (β, w_f w_0(β'))=(β, w_f(β')); a reindexing w_f → w_f^{-1} then yields the expression used in (4.4). This repair should be written out explicitly. Because (4.4) is the pivotal identity that identifies the S-matrix with that of W(sl_s[s], s/u), the step is load-bearing for Theorem 4.12.","section":"§4.3.1, Proposition 4.20"},{"comment":"The paper states in Section 2.4 that the minimal (most negative) conformal dimension is attained on a unique irreducible module, that this is 'believed to be true in general, and is verified in the cases treated in this paper (see Section 4.3.2)'. However, Section 4.3.2 only proves m-independence of the list of conformal dimensions (Corollary 4.26) and m-independence of the central charge (Proposition 4.27). It does not prove uniqueness of the minimum. This uniqueness is used in equation (2.3) to guarantee positivity of S_{i,◦}/S_{1,◦}, and Lemma 2.11 uses that positivity to eliminate the sign factors in the final S-matrix comparison. Without an explicit verification of uniqueness (or an alternative sign-fixing argument not relying on it), the conclusion of Theorem 4.12 is not fully established. This is a load-bearing gap and should be addressed.","section":"§2.4 and §4.3.2"}],"minor_comments":[{"comment":"The level in Theorem 1.1 and in the abstract is written as W(sl_n[u^m,s], n/s). From the rest of the paper (e.g. Proposition 4.20, Proposition 4.27, Section 4.3.2) the intended level is clearly n/u, not n/s. Please correct this typo in the theorem statement.","section":"Theorem 1.1 and Abstract"},{"comment":"The text says 'in order to apply Proposition 2.11', but the statement used is Lemma 2.11. Please fix the cross-reference.","section":"§4.3.1, after (4.6)"},{"comment":"In the use of Lemma 4.15 to average over W_0, the constant |W_0| arising from the identity is absorbed into C(λ,λ') without comment. This is harmless but should be stated for clarity.","section":"§4.3.1, proof of Proposition 4.16"},{"comment":"For the E_8 subregular example, the choices of y ∈ W giving the listed β are said to exist but are not recorded. Since the numerical S-matrix data rely on these choices, a remark on how they are obtained (or a reference to a computational appendix) would improve reproducibility.","section":"§3.4"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely correct in its main claims, and the two major issues are repairable rather than fatal: Proposition 4.20 can be fixed by a short rewriting, and the uniqueness of the minimal conformal dimension should be verified from formula (4.12) or a known result. The manuscript's fit with math.QA is excellent. I would ask the authors to make these repairs and to correct the n/s vs n/u typo before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a substantial paper with a genuinely new result: equality of q-characters and modular data between boundary W-algebras in type A and principal W-algebras. That equality, Theorem 1.1, and the fusion-rule corollary go beyond the conjectures in [8,47] and give the first provable statements of this kind. The paper deserves a serious referee, but not acceptance as it stands.\n\nWhat is new and good: the q-character equality (Theorem 4.8) is proven by a clean combinatorial induction on m; the Type D product-formula explanation in Section 3.2 is elegant and correct; and the E8 subregular modular data is a valuable concrete computation. The overall architecture rests on established theorems—Kac–Wakimoto characters, rationality, Huang's Verlinde theorem—rather than on the claims being proved.\n\nThe critical soft spot is Proposition 4.20. After restricting the S-matrix sum to W^f W_0, the proof replaces (β, w_f w_0 β') with (β, w_f β'), citing (β, Δ_{0,+})=0. That is not valid: w_0 β' − β' lies in the span of Δ_0, but the inner product is with w_f^{-1}β, not with β. The needed orthogonality (w_f^{-1}β, Δ_0)=0 is not established and is generally false in type A pyramid geometry (the n=11, u=8, s=3 case already shows it). The product factor does not rescue the step, because the w_0-sum also carries the exponential. So formula (4.4) is not derived, and since (4.4) is the pivot for comparing to the principal W(sl_s) S-matrix, Theorem 4.12 and the fusion-rule corollary rest on an unproved step. The result may well be true and repairable—the surrounding arguments are careful—but the gap is real and load-bearing.\n\nSecondary issues: Section 2.4 asserts a unique minimal conformal dimension module, 'verified in Section 4.3.2,' but that section only proves m-independence, not uniqueness. Lemma 2.11 and equation (2.3) rely on positivity anchored at that unique minimal module, so the sign-fixing needs an explicit argument. The E8 computation also ships no code; the listed minimal polynomials and numerical values help, but full reproducibility is not available.\n\nWho this is for: vertex algebra and modular tensor category people, especially those working on W-algebras and fusion rules. I would send it to a serious referee, and I would want the revision to fix Proposition 4.20 and the minimal-dimension point before I trust Theorem 4.12.","headline":"Substantial new results on boundary W-algebras, but Proposition 4.20 has a load-bearing gap that the paper does not address; the main theorem is not proven as written.","tokens_in":29362,"tokens_out":6661,"would_cite":true,"duration_ms":70216,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["17B69","17B67"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that two families of exceptional W-algebras in type A—a boundary algebra built from sl_n and a principal algebra built from sl_s—have the same irreducible modules, the same q-characters, and the same modular data, hence the","keywords":["exceptional W-algebras","boundary W-algebras","fusion rules","q-characters","modular data","vertex algebras","type A nilpotent orbits","necklace combinatorics"],"falsifier":"List the conformal dimensions given by formula (4.12) for all irreducible modules of W(sl_7[5,2], 7/2) (u=5, s=2, m=1) and check whether the minimum occurs exactly once. If two distinct modules have the same minimal conformal dimension, then the sign-fixing step relying on equation (2.3) and Lemma 2.11 fails to apply, and Theorem 1.1's equality of S-matrices is not established by the paper's argument. A second check: compute S-matrix entries directly for a small pair (u,s) via the determinant reduction of Section 3.4 and compare them with the S-matrix of the principal W-algebra W(sl_s[s], s/u)","tokens_in":28386,"feed_emoji":"🧩","tokens_out":10704,"duration_ms":114462,"temperature":0.7,"pith_summary":"This paper proves a matching theorem for two families of vertex algebras (the algebraic structures underlying two-dimensional conformal field theory) built from the Lie algebras sl_n and sl_s. For coprime u and s, with n = mu + s, the exceptional W-algebra W(sl_n[um,s], n/u) and the principal W-algebra W(sl_s[s], s/u) are shown to have the same list of irreducible modules, the same q-characters (infinite-product formulas for the graded traces), and the same modular data. Since modular data determine fusion rules, the boundary algebras inherit the fusion rules of the principal algebras, which were already known. The paper then uses a factorization phenomenon to extend this from boundary levels to all exceptional W-algebras in type A, giving a largely complete determination of their fusion rules. The result matters because it reduces a large part of W-algebra representation theory to a single combinatorial parameter—necklaces—that is independent of the integer m.","feed_headline":"Boundary and principal W-algebras match on characters and fusion rules","feed_subtitle":"Boundary algebras built from sl_n match principal algebras from sl_s, pinning down type A exceptional fusion rules.","key_machinery":"The key objects are the boundary W-algebra W(sl_n[um,s], n/s) and the principal W-algebra W(sl_s[s], s/u), together with a necklace bijection between their parameter sets P^u_{+,f}: irreducible modules are indexed by circular arrangements of s long blocks and u−s short blocks, whose count is (1/u) C(u,s), independent of m. Three identities carry the argument: (1) the q-character product formula reduces character equality to m-independence of dim(m,k) + dim(m,u−k) − dim(g0); (2) the S-matrix sum over W(Γ) is reduced to a sum over the row subgroup W^f and then factors into a W^(s)-component times a phase, with the phase shown to be C_0 ε(β)ε(β′) via the Killing-form identity ⟨β,β′⟩ = (1/2)κ_{g","core_discovery":"The central claim is Theorem 1.1 (with the fusion-rule version in Theorem 4.12): for n = mu + s, 1 ≤ s ≤ u − 1, gcd(s,u) = 1, the exceptional W-algebra W(sl_n[um,s], n/s) and the principal W-algebra W(sl_s[s], s/u) have irreducible modules in natural bijection, equal q-characters under that bijection, and equal modular data. The proof gives a concrete bijection: irreducible modules of both algebras are parameterized by u-bead necklaces with s long blocks—(1/u) C(u,s) of them—and the same necklace parameter appears on both sides. Equality of characters follows from a dimension-counting identity for pyramids; equality of S-matrices follows from reducing a Weyl-group sum to a sum over a row sub","pith_inferences":["Beyond the paper: if the equality of modular data could be upgraded to an isomorphism of vertex algebras, the boundary and principal families would be the same object presented in different ways; the paper explicitly leaves this open, noting that equal modular data alone is not enough to force isomorphism.","Beyond the paper: the necklace parameter set is independent of m, suggesting the entire fusion category of W(sl_{mu+s}[um,s], n/u) stabilizes as m varies; a direct check would be to compute fusion coefficients for two different m values with the same u and s and compare them term by term.","Beyond the paper: the type-D application in Section 3.2 explains certain product formulas for Virasoro minimal-model characters as boundary W-algebra characters; the same mechanism likely produces analogous product formulas for other W-algebra families, and searching for them at fixed u with varying m would be a natural test."],"forward_implications":["All boundary W-algebras in type A have their fusion rules completely determined: they coincide with those of the principal W-algebra W(sl_s[s], s/u), whose fusion rules are already known.","For odd u and s, the Grothendieck ring of every exceptional W-algebra W(sl_n[um,s], p/u) factors as F(L_{u−s}(sl_s))^{int} ⊗ F(L_{p−n}(sl_n)), giving a largely complete determination of exceptional fusion rules in type A.","The q-character equality gives an explicit infinite-product formula for every irreducible module of the boundary W-algebra, with all dependence on the module encoded in a necklace parameter.","The new S-matrix formula yields explicit modular data in cases outside the boundary family, including a 44-module modular tensor category for the E8 subregular W-algebra at denominator 29."],"fun_headline_variants":["Boundary W-algebras equal principal ones: characters and fusion","Fusion rules of exceptional W-algebras from a boundary-principal match","Same necklaces yield same characters and fusion rules in W-algebras","Boundary-principal W-algebra agreement: q-characters and fusion","Exceptional W-algebra fusion rules via boundary equality"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the minimal conformal dimension is attained by a unique irreducible module; Section 2.4 says this is believed true in general and verified in the cases treated in the paper, but Section 4.3.2 proves m-independence of conformal dimensions and central charge, not uniqueness. If two distinct modules tie for the minimum, the positivity of quantum dimensions and the sign-fixing step (Lemma 2.11) would require an additional argument, and the equalit","fun_headline_variants_meta":{"raw":{"variants":["Boundary W-algebras equal principal ones: characters and fusion","Fusion rules of exceptional W-algebras from a boundary-principal match","Same necklaces yield same characters and fusion rules in W-algebras","Boundary-principal W-algebra agreement: q-characters and fusion","Exceptional W-algebra fusion rules via boundary equality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001227,"raw_usage":{"total_tokens":4806,"prompt_tokens":599,"completion_tokens":4207,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":343,"completion_tokens_details":{"reasoning_tokens":4115}},"tokens_in":343,"tokens_out":4207,"duration_ms":37123,"temperature":1.0,"reasoning_tokens":4115,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T19:47:59.485419+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"List the conformal dimensions given by formula (4.12) for all irreducible modules of W(sl_7[5,2], 7/2) (u=5, s=2, m=1) and check whether the minimum occurs exactly once. If two distinct modules have the same minimal conformal dimension, then the sign-fixing step relying on equation (2.3) and Lemma 2.11 fails to apply, and Theorem 1.1's equality of S-matrices is not established by the paper's argument. A second check: compute S-matrix entries directly for a small pair (u,s) via the determinant reduction of Section 3.4 and compare them with the S-matrix of the principal W-algebra W(sl_s[s], s/u)","supporting_citations":[],"review_version":1}