{"id":"cc003b18-6517-4c2f-8046-8e7618ed6c5a","arxiv_id":"2501.00401","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"For any tensor product of irreducible polynomial gl(m|n)-modules, the Gaudin algebra on the singular space is cyclic, Frobenius, and generically diagonalizable with simple spectrum, yielding a complete reformulated Bethe ansatz.","lead":"This paper proves that the Gaudin algebra of the general linear Lie superalgebra gl(m|n) acts on tensor products of polynomial modules with a single cyclic generator and with simple spectrum for generic parameters. The result gives a complete diagonalization of the super Gaudin model, expressed through Fuchsian differential operators.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified.","rationale":"The central claim is that the Gaudin algebra of gl_{m|n} acts cyclically on the singular space of any tensor product of irreducible polynomial modules, with a Frobenius quotient, and is generically diagonalizable with simple spectrum. The proof reduces to the known non-super case for gl_{m+r} via the Berezinian identities of Proposition 3.16. I checked this reduction in detail. The apparent worry that E_{i,m}(z) need not annihilate a singular vector is resolved by the nonnegativity of weights: if the total weight has zero m-th coordinate, then every tensor factor has zero m-th coordinate, so no monomial contains e_m and each local operator E_{i,m}^{(k)} kills the vector. The Berezinian calculation in part (ii) is also correct: for an odd last block, the Schur complement contributes its inverse, so equation (3.8) is the proper rearrangement of Proposition 2.2. The proof of Theorem 4.7 then goes through: the lifted weights have vanishing odd part by the choice of r, Proposition 3.16 identifies the actions of the relevant Gaudin algebras, and the eigenbasis transfer in Proposition 4.9 correctly shifts the differential operator by powers of ∂_z^{-1}. The only blemish is the terse justification that a direct summand of a cyclic B_{m+r}(z)-module is cyclic. That statement is not true for arbitrary modules, but in this setting the full singular space is free of rank one over the Gaudin algebra because of the Frobenius property established in Remark 4.5, so every direct summand is cyclic. This is a minor expository gap rather than a flaw in the mathematics. The reader's identification of Proposition 3.16 as the most fragile link is reasonable, but my independent check found the calculation sound; hence no adjustment to the ACCEPT verdict is needed.","tokens_in":24664,"tokens_out":49329,"duration_ms":452398,"concrete_test":"Verify Proposition 3.16(i) in the non-super case gl_3 with p=2 on the singular weight space of weight (2,1,0) inside (C^3)^{⊗3}, using a singular vector that is not a product of highest weight vectors of the tensor factors. Compute the action of Ber(L_{3}(z)) = cdet(L_3(z)) on that vector and compare with the action of cdet(L_2(z)) ∂_z; if the two agree as formal pseudo-differential operators, the key reduction survives this nontrivial case.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I attempted to break the two most load-bearing steps and found no mathematical flaw. Proposition 3.16, identified by the reader as the weakest assumption, survives scrutiny: the claim that E_{i,m}(z) acts trivially on the relevant singular weight space is justified by the nonnegativity of polynomial weights together with Lemma 3.9, and the Berezinian block computation in part (ii) correctly uses the inverse of the odd Schur complement, so equation (3.8) is valid. The reduction in Theorem 4.7, via the lift to gl_{m+r|n} and truncation to gl_{m+r}, is also sound: the choice of r makes the odd part of the lifted weights vanish, so Proposition 3.16 applies. The only presentational gap is in Theorem 4.7(i), where a direct summand N of a cyclic B_{m+r}(z)-module is asserted to be cyclic because B_{m+r} preserves singular weight spaces. That reason alone is insufficient in general, but here the conclusion follows from the already-established Frobenius property (Remark 4.5 plus Lemma 4.2), which makes the full singular space free of rank one over the Gaudin algebra, so every weight-space direct summand is cyclic. This is a repairable expository omission, not a substantive error.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the Gaudin algebra B_{m|n}(z) of the general linear Lie superalgebra gl_{m|n} acting on tensor products of irreducible polynomial modules. The main results are Theorem 4.7, asserting that the singular space M^sing is a cyclic B_{m|n}(z)-module and that the image of B_{m|n}(z) in End(M^sing) is a Frobenius algebra, and Theorem 4.10, asserting diagonalizability with a simple spectrum for generic z. The proof proceeds by lifting M to a gl_{m+r|n}-module, truncating to gl_{m+r}, and transferring the known cyclicity and diagonalizability results for gl_{m+r} down to gl_{m|n} using the Berezinian calculus and the key identity Proposition 3.16. The final section relates the diagonalization to Bethe vectors and Fuchsian differential operators with polynomial kernels, yielding a completeness statement for a reformulated Bethe ansatz.","tokens_in":32,"tokens_out":3219,"duration_ms":154861,"significance":"If correct, the paper establishes a substantial super analogue of the Gaudin Bethe ansatz completeness theorem, extending the results of Rybnikov and Mukhin-Tarasov-Varchenko from gl_m to gl_{m|n}. The reduction strategy is elegant: rather than developing a new superalgebraic diagonalization method, the authors transfer the known non-super results through a carefully chosen odd reflection and truncation. The key technical identity, Proposition 3.16, is a direct Berezinian calculation, and the stress-test scrutiny found it correct; this lends credibility to the main theorems. The paper also gives explicit connections to Bethe vectors and Fuchsian differential operators, and it identifies a plausible conjecture about the Feigin-Frenkel center, which adds context. The proofs rely on published external results rather than circular reasoning, and the manuscript is careful in citing the sources for the structural facts it uses.","major_comments":[],"minor_comments":[{"comment":"The sentence 'as B_{m+r} preserves singular weight spaces' does not by itself imply that the direct summand N of a cyclic module is cyclic; this implication is false in general. The conclusion nevertheless follows from the already established Frobenius property of the Gaudin algebra on the full singular space (Remark 4.5) together with Lemma 4.2(iii), which implies that every weight-space direct summand is cyclic. Please replace the given reason with this argument.","section":"Section 4.3, proof of Theorem 4.7(i)"},{"comment":"The equality N := ⊕_μ tr_{p|n}(M)^sing_μ = ⊕_μ M^{σ_p-sing}_μ is slightly ambiguous because the two direct sums are indexed by different sets of weights; the first is over singular weights of tr_{p|n}(M) and the second over σ_p-singular weights of M. The intended identification via Corollary 3.13 should be stated more explicitly.","section":"Section 3.4, Proposition 3.16"},{"comment":"In the statement of Theorem 5.4, the notation Δ_{η,ᵇ́γ,z} is used without repeating the definitions of η and ᵇ́γ; the surrounding text defines them, but restating them in the theorem statement would improve readability.","section":"Section 5.2, Theorem 5.4"},{"comment":"There is a typo: '[MR, Corallary 3.6]' should read '[MR, Corollary 3.6]'.","section":"Section 3.2"},{"comment":"The word 'isomophism' should be 'isomorphism'.","section":"Section 4.3, proof of Theorem 4.7"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this paper proves the right thing and the proof hangs together. The main results—cyclicity, Frobenius property, and generic diagonalizability with simple spectrum for the Gaudin algebra B_{m|n}(z) on the singular space of tensor products of irreducible polynomial gl(m|n)-modules—are new and not in the cited literature. Prior work covered gl(m) (Rybnikov, Mukhin–Tarasov–Varchenko), the cubic Hamiltonians in [CCL], and gl(1|1) (Lu). The paper closes the gap for all gl(m|n).\n\nThe proof strategy is the interesting part: lift to gl(m+r|n), use odd reflections to convert a singular weight space into a sigma_m-singular space with zero weights on the last r even coordinates, then reduce to the non-super gl(m+r) Gaudin algebra via the Berezinian identity in Proposition 3.16. That identity is the load-bearing step, and it is a direct block-elimination calculation. I checked the two places a skeptical reader might distrust: the claim that E_{i,m}(z) acts trivially on the relevant weight space (justified by polynomial weights plus Lemma 3.9), and the inverse Schur complement in part (ii). Both hold. The stress-test note agrees, and I think it is right.\n\nThe soft spots are minor. The proof of Theorem 4.7(i) says a direct summand of a cyclic B_{m+r}(z)-module is cyclic because the algebra preserves singular weight spaces. That implication is false in general. Here the conclusion still follows because the Frobenius property (already established via Lemma 4.2 and the Shapovalov form) makes the full singular space free of rank one over the Gaudin algebra, so each weight-space summand is cyclic. This is an expository gap, not a gap in the mathematics, and it should be patched.\n\nThe word \"generic\" in Theorem 4.10 is inherited from the non-super diagonalization theorem; the paper does not quantify it. That is standard but could be said more explicitly. The final section on the Feigin–Frenkel center is speculative but clearly marked as such, and it does not affect the main results.\n\nCitation pattern: the paper leans on published results (Rybnikov, Mukhin–Tarasov–Varchenko, Molev–Ragoucy, Cheng–Lam), and the self-citations are for background lemmas that are available elsewhere. No red flag.\n\nWho it is for: anyone working on Gaudin models, Bethe ansatz, or superintegrable systems. It deserves a serious referee; I would send it out and expect a quick accept after minor revision.","headline":"Main results are new and the reduction to the non-super Gaudin algebra is sound; worth sending to a strong referee with a minor request to fix the cyclicity argument in Theorem 4.7(i).","tokens_in":25446,"tokens_out":2167,"would_cite":true,"duration_ms":20049,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["17B10","17B81","81R12"],"pacs":[],"model":"deepseek-v4-flash","headline":"The Gaudin algebra of gl(m|n) is cyclic, maximal, and diagonalizable on singular spaces.","keywords":["Gaudin model","Lie superalgebra gl(m|n)","Berezinian","Bethe ansatz completeness","Frobenius algebra","Fuchsian differential operators","polynomial modules","odd reflections"],"falsifier":"In the smallest nontrivial super case, $\\mathfrak{gl}(2|1)$ with two sites and fixed polynomial tensor factors, compute the common eigenvectors of the Gaudin Hamiltonians at a generic pair $(z_1,z_2)$; if their number is less than $\\dim M^{\\mathrm{sing}}$, or if some eigenspace has dimension greater than one, the theorem fails. A cheaper check is the reduction identity itself: on a $\\sigma_2$-singular weight space with vanishing third-even weight, verify that $\\operatorname{Ber}(L_{3|1}(z))v = \\operatorname{Ber}(L_{2|1}(z))\\partial_z v$ for every $v$ in that space.","tokens_in":24439,"feed_emoji":"🧮","tokens_out":11684,"duration_ms":104250,"temperature":0.7,"pith_summary":"The paper proves that the Gaudin algebra of the general linear Lie superalgebra $\\mathfrak{gl}_{m|n}$ completely controls the singular space of any tensor product of irreducible polynomial modules: that singular space is generated from a single vector, and for generic distinct site parameters the algebra acts with one-dimensional eigenspaces. This gives a superalgebra analogue of the strongest completeness statements known for ordinary Gaudin models, where the Bethe ansatz is replaced by a separation-of-variables description through differential operators. The proof embeds the super Gaudin model into a larger ordinary general-linear Gaudin model, then uses odd reflections and a Berezinian reduction identity to transfer cyclicity and diagonalizability downward. If correct, the Gaudin algebra on the singular space has dimension equal to that space and is a maximal commutative subalgebra of its endomorphisms.","feed_headline":"Super Gaudin model: complete diagonalization on singular space","feed_subtitle":"Tensor products of polynomial gl(m|n)-modules get one-dimensional eigenspaces, a super version of Bethe completeness.","key_machinery":"The load-bearing object is the Berezinian of the Lax matrix $L_{m|n}(z)$, a $(m+n)\\times(m+n)$ matrix of super pseudo-differential operators whose coefficients generate the Gaudin algebra $\\mathcal{B}_{m|n}(z)$. The Berezinian is the super-analogue of the determinant that makes the transfer-matrix construction commutative; the matrix is of Manin type, meaning its entries satisfy the super-commutation relations that make the Berezinian well behaved. The transfer step is Proposition 3.16: on a $\\sigma_p$-singular weight space whose last $m-p$ even diagonal weights vanish, the Berezinian of $L_{m|n}(z)$ acts exactly like the Berezinian of the smaller matrix $L_{p|n}(z)$ followed by a power of the formal derivative $\\partial_z$, and a dual statement truncates odd coordinates. Odd reflections — changes of Borel subalgebra that permute the ordering of even and odd roots — provide isomorphisms between the relevant singular weight spaces, so the known cyclicity and diagonalizability of ordinary $\\mathfrak{gl}_{m+r}$ Gaudin algebras can be transported down to $\\mathfrak{gl}_{m|n}$.","core_discovery":"The central claim, stated as Theorem 4.7 and Theorem 4.10, is that for any pairwise distinct complex parameters $z=(z_1,\\ldots,z_\\ell)$ and any $\\ell$-fold tensor product $M$ of irreducible polynomial $\\mathfrak{gl}_{m|n}$-modules, the singular space $M^{\\mathrm{sing}}$ is cyclic as a module over the Gaudin algebra $\\mathcal{B}_{m|n}(z)$, and the image of $\\mathcal{B}_{m|n}(z)$ on $M^{\\mathrm{sing}}$ is a Frobenius algebra. Consequently this image has dimension $\\dim M^{\\mathrm{sing}}$, is a maximal commutative subalgebra of $\\operatorname{End}(M^{\\mathrm{sing}})$, and every eigenspace is one-dimensional; for generic $z$ the algebra is diagonalizable with a simple spectrum. The paper also constructs an eigenbasis: it is pulled back, through an odd-reflection isomorphism, from the known eigenbasis of an ordinary $\\mathfrak{gl}_{m+r}$ Gaudin algebra, and the corresponding eigenvalues are encoded by monic Fuchsian differential operators with polynomial kernels. The authors state that this should be read as the completeness of a reformulation of the Bethe ansatz for $\\mathcal{B}_{m|n}(z)$ acting on $M^{\\mathrm{sing}}$.","pith_inferences":["The reduction identity in Proposition 3.16 suggests the completeness statement may extend beyond polynomial modules, provided the singular weights satisfy the same vanishing condition; testing the smallest non-polynomial case, such as a $\\mathfrak{gl}(1|1)$ factor with a non-polynomial highest weight, would separate the hook-partition assumption from the Berezinian mechanism.","One can test Theorem 4.10 numerically without building the Gaudin algebra: for small $m,n$, count the monic Fuchsian operators with polynomial kernels and prescribed exponents at the marked points, and compare that count with $\\dim M^{\\mathrm{sing}}$ at a generic $z$; equality is exactly the paper's picture.","The maximality of the image on $M^{\\mathrm{sing}}$ is evidence for the paper's Conjecture 5.5 that the full affine vertex algebra center is generated by derivatives of the Berezinian coefficients, because a larger center would still act through the same maximal commutative algebra on singular spaces."],"forward_implications":["The Gaudin algebra image on $M^{\\mathrm{sing}}$ has dimension exactly $\\dim M^{\\mathrm{sing}}$ and is a maximal commutative subalgebra of $\\operatorname{End}(M^{\\mathrm{sing}})$.","Every eigenspace of the Gaudin algebra on $M^{\\mathrm{sing}}$ is one-dimensional, so the Gaudin Hamiltonians have a joint spectrum with no accidental degeneracy.","For generic parameters the algebra is diagonalizable with a simple spectrum, giving a simultaneous eigenbasis for all Gaudin Hamiltonians on $M^{\\mathrm{sing}}$.","An eigenbasis can be obtained from Fuchsian differential operators of appropriate order with polynomial kernels, and the eigenvalues are read off from their coefficients; this is a super version of the geometric Langlands correspondence for Gaudin models.","If the ordinary Bethe ansatz is complete for the embedded $\\mathfrak{gl}_{m+r}$ model, then Bethe vectors give an explicit eigenbasis for the super Gaudin algebra as well."],"supporting_citations":[{"why":"supplies the base theorem that the singular space of a $\\mathfrak{gl}_m$ tensor product is cyclic over the ordinary Gaudin algebra.","marker":"[Ry]"},{"why":"establishes the ordinary $\\mathfrak{gl}_m$ diagonalization with simple spectrum and the Fuchsian-operator eigenbasis that the super case is pulled back from.","marker":"[MTV3]"},{"why":"proves the separation-of-variables completeness result for the quantum $\\mathfrak{gl}_N$ Gaudin model used as the non-super input.","marker":"[MTV4]"},{"why":"constructs the $\\mathfrak{gl}_{m|n}$ Gaudin algebra from the affine vertex algebra center via Berezinians and proves its commutativity.","marker":"[MR]"},{"why":"provides the Manin-matrix and Berezinian identities, including the block factorizations behind Proposition 3.16.","marker":"[HM]"},{"why":"gives the odd-reflection identification of $\\sigma$-singular vectors used in the truncation isomorphisms.","marker":"[CL]"},{"why":"supplies the lemmas converting cyclicity into Frobenius-algebra structure, maximality, and one-dimensional eigenspaces.","marker":"[Lu1]"},{"why":"provides the Bethe vectors and the eigenvalue formula for the ordinary Gaudin algebra used in the super eigenvalue description.","marker":"[MTV1]"}],"fun_headline_variants":["Gaudin superalgebra diagonalized on singular space","Bethe ansatz completeness for gl(m|n) Gaudin","Super Gaudin: simple spectrum on tensor products","Frobenius Gaudin algebra yields full eigenbasis","Gaudin model for gl(m|n): completeness of Bethe"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is a block-matrix calculation: on a singular weight space whose last even coordinate weights vanish, the Berezinian of the large Gaudin Lax matrix acts exactly like the Berezinian of a smaller matrix composed with a power of the formal derivative; if that equality fails, the transfer argument from the known ordinary-linear case collapses.","fun_headline_variants_meta":{"raw":{"variants":["Gaudin superalgebra diagonalized on singular space","Bethe ansatz completeness for gl(m|n) Gaudin","Super Gaudin: simple spectrum on tensor products","Frobenius Gaudin algebra yields full eigenbasis","Gaudin model for gl(m|n): completeness of Bethe"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000581,"raw_usage":{"total_tokens":2794,"prompt_tokens":1061,"completion_tokens":1733,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":677,"completion_tokens_details":{"reasoning_tokens":1650}},"tokens_in":677,"tokens_out":1733,"duration_ms":12573,"temperature":1.0,"reasoning_tokens":1650,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:52:06.967649+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In the smallest nontrivial super case, $\\mathfrak{gl}(2|1)$ with two sites and fixed polynomial tensor factors, compute the common eigenvectors of the Gaudin Hamiltonians at a generic pair $(z_1,z_2)$; if their number is less than $\\dim M^{\\mathrm{sing}}$, or if some eigenspace has dimension greater than one, the theorem fails. A cheaper check is the reduction identity itself: on a $\\sigma_2$-singular weight space with vanishing third-even weight, verify that $\\operatorname{Ber}(L_{3|1}(z))v = \\operatorname{Ber}(L_{2|1}(z))\\partial_z v$ for every $v$ in that space.","supporting_citations":[],"review_version":1}