{"id":"fba9425a-4ec2-48d0-9e9c-63d0e67cc03e","arxiv_id":"1908.07809","paper_version":2,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"For every tame extended affine Lie algebra with a reduced root system, a group is built whose subgroup quotient recovers the extended affine Weyl group.","lead":"The authors construct Steinberg groups and Kac-Peterson integration groups for tame extended affine Lie algebras with reduced root systems, extending earlier work that covered only simply-laced types. They then show that the extended affine Weyl group appears as a quotient of two subgroups inside the constructed group, paralleling the Kac-Moody group story.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Consequence 4.4 is unproved and, for isotropic root sums, false in affine cases; the homomorphism χ in Proposition 4.7 is therefore not established and Theorem 4.23 is unsupported.","rationale":"The paper's strategy—define a Steinberg group for an EARS, map it to Kac-Peterson integration groups, and recover the extended affine Weyl group as Ad(N)/Ad(T)—is natural, and the result would be a plausible generalization of [16]. However, the proof as written has a central gap at Consequence 4.4. St2 is a one-way relation: it only specifies commutators for nilpotent pairs α+β ∈ R^×. Consequence 4.4 treats the converse as if it were a defining relation, but in a group presentation the absence of a relation cannot imply commutativity. The proof's stated reason, R+_{α,β} = ∅, is also incorrect when α+β is an isotropic root: in affine type A1^(1), α + (-α+δ) = δ while 2α + (-α+δ) = α+δ is a real root. In standard affine Kac-Moody groups such root subgroups do not commute, so the consequence contradicts the paper's own Remark 4.5. This gap is load-bearing because Proposition 4.7's χ, defined via the unordered product (4.14), is used to establish the epimorphisms in Proposition 4.12, and Theorem 4.23 uses those epimorphisms to lift the generalized presentation of the Weyl group. If the Steinberg relations were repaired by adding explicit commutator relations for pairs with isotropic sums, and if χ were re-proved with those relations, the main theorem might be salvageable, but that is not accomplished in the manuscript.","tokens_in":17492,"tokens_out":15361,"duration_ms":164135,"concrete_test":"In the affine Kac-Moody group of type A1^(1), compute the commutator [xα(t), x_{-α+δ}(s)] using either the level-one integrable representation of sl2(C[t,t^-1])⊕Cc⊕Cd or the standard Tits presentation. If the commutator has a nontrivial x_{α+δ}(c t^2 s)-component, Consequence 4.4 is false. Then check the map χ of Proposition 4.7 on the relation [xα(a), x_{-α}(b)] in St_{A1}(Cσ): its image is a product of commutators of root subgroups x_{α+δ}(tδ) and x_{-α+τ}(sτ); if the δ+τ ≠ 0 terms do not commute, χ is not a homomorphism.","verdict_should_be":"REJECT","load_bearing_attack":"Definition 4.3, St2 (equation 4.7), imposes a commutator formula only when α+β ∈ R^× (a nilpotent pair). Consequence 4.4 asserts the converse: if α+β ∉ R^× then (xα(t),xβ(s)) = 1, with the explanation that R+_{α,β} is empty. Neither step is valid. An omitted relation is not a trivial relation, and the claimed emptiness is false: in an affine A1^(1) subsystem, take α = α and β = -α + δ, where δ is isotropic. Then α+β = δ ∈ R0, so not in R^×, but 2α+β = α+δ ∈ R^×; hence R+_{α,β} is nonempty. In the standard affine Kac-Moody group, which Remark 4.5 says this construction should specialize to, [xα(t), x_{-α+δ}(s)] is not trivial; the Tits commutator formula for this pair has a nonzero x_{α+δ}(c t^2 s)-term. Consequence 4.4 is used in Proposition 4.7 to make the product defining χ in (4.14) well-defined and to verify that χ is a homomorphism. Since the consequence is unsupported and false in the affine case, the map χ from St_{˙R}(Cσ) to St_{R,˙R,σ}(C) need not exist. Proposition 4.12 and the final Weyl-group isomorphism in Theorem 4.23 both depend on this χ, so the central argument fails as written.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a construction of Steinberg groups associated to reduced extended affine root systems (EARS), introduces extended affine Kac–Moody groups by generators and relations, and then uses the Kac–Peterson integration method to attach a group to every tame extended affine Lie algebra (EALA). The main theorem states that the extended affine Weyl group W of such an EALA is isomorphic to Ad(N)/Ad(T), where N and T are the subgroups of the integrated group generated by the elements n_α(t) and h_α(t) for non-isotropic roots α, thereby generalizing Kryliouk's simply-laced result to all reduced EARS. The argument relies on a presentation of W by conjugation from [7], on a map χ from the finite-type Steinberg group over a quantum torus to the newly defined Steinberg group, and on several structural results about EALAs and their root subsystems.","tokens_in":17825,"tokens_out":6011,"duration_ms":154531,"significance":"If the central claims were correct, the paper would be a substantial contribution: it would provide a uniform Steinberg-group construction for arbitrary reduced extended affine root systems, an integration functor for tame EALAs, and a group-theoretic realization of the extended affine Weyl group as a quotient of two natural subgroups. The paper also makes useful connections with earlier work of Kryliouk and with the generalized presentation by conjugation of Azam. However, the main construction depends on a commutativity assertion that is false in the affine case, and several load-bearing proofs are only sketched. As written, the central theorem is not established.","major_comments":[{"comment":"Consequence 4.4 is false as stated. The commutator relation St2 is imposed only when α+β ∈ R^×, so it does not imply commutativity when α+β is isotropic; an omitted relation is not a trivial relation. The asserted emptiness of R^+_{α,β} is also incorrect: for roots α and β = -α+δ with δ ∈ R0 and α+δ ∈ R^×, one has α+β = δ ∉ R^× but 2α+β = α+δ ∈ R^×, so R^+_{α,β} is nonempty. This configuration occurs inside an A_1^(1) subsystem. In the affine Kac-Moody Steinberg group, which Remark 4.5 says the present construction should specialize to, the commutator [x_α(t), x_{-α+δ}(s)] has a nontrivial x_{α+δ}(c t^2 s) term. Since Consequence 4.4 is used in Proposition 4.7 to make the product in (4.14) independent of the order of factors and to verify that χ is a homomorphism, the map χ from St_{\\dot R}(Cσ) to St_{R,\\dot R,σ}(C) is not established. Proposition 4.12 and Theorem 4.23 both depend on this map, so the central argument fails as written.","section":"Section 4.1, Consequence 4.4 and Definition 4.3 (Eq. (4.7))"},{"comment":"The four isomorphisms Gnil(C) ≅ Gnil,c(C) ≅ Gint(C) ≅ Gint,c(C) are asserted with a reference to [16, Proposition 3.2.41] and a remark that the proof is similar, but the paper gives no indication of which parts of the simply-laced proof adapt to arbitrary reduced EARS or what new difficulties arise from non-simply-laced types and nontrivial 2-cocycles. This is load-bearing because Theorem 4.23 defines Ad(G) after these identifications, so the isomorphism class of \\bar G is essential. A complete proof or a precise statement of the adapted argument is required.","section":"Section 4.2, Proposition 4.22"},{"comment":"The verification that the reduced-collection relations of Theorem 2.3(iii) are respected is too compressed. Equation (4.41) rewrites \\tilde c(α,δ) as a product of \\hat n_{\\dot α} factors, and (4.42) rewrites it in terms of \\hat h factors, but the paper does not show that a product over a reduced collection with its signs and exponents maps to the identity in Ad(N)/Ad(T), nor how the contributions of the coefficients η_p are controlled. Without this, the induced homomorphism from the presented group \\hat W is not established.","section":"Theorem 4.23, relation (iii), Eqs. (4.41)–(4.42)"}],"minor_comments":[{"comment":"There are numerous typographical errors, including \"emphesize\" for \"emphasize\", \"underlined filed\" for \"underlying field\", \"without loose of generality\" in Proposition 3.8, and \"Rα,b\" for \"Rα,β\" in the proof of Proposition 4.12. These should be corrected.","section":"Throughout"},{"comment":"The phrase \"preserves T orin\" should be \"preserves the T-or relation\" from Definition 4.10.","section":"Consequence 4.15"},{"comment":"In relation (ii), the notation \\hat r_{r_α(β)} is unclear; the subscript should be written as \\hat r_{r_α(β)} or \\hat r_{w_α(β)} to denote the generator indexed by the reflected root.","section":"Theorem 2.3"},{"comment":"Reference [16] is a PhD thesis; the citations to [16, Proposition 3.2.41] and to \"Page 130\" would benefit from more precise statements, since the reader cannot easily verify the adapted claims without the thesis in hand.","section":"References"}],"recommendation":"reject","confidential_remarks":"The central problem is not a minor gap: Consequence 4.4 is false in the affine case and contradicts the paper's own Remark 4.5 that the construction specializes to affine Kac-Moody groups. Because the map χ in Proposition 4.7 and the subsequent epimorphism to the integrated group depend on this consequence, a local revision cannot repair Theorem 4.23 without changing the Steinberg presentation itself. I therefore recommend rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nLet me save you time: this paper is aiming at a real generalization, but there is a load-bearing gap in the written proof. Consequence 4.4 is not derived and, in the affine case the paper says it should cover, it is false. The map χ in Prop 4.7 and everything after (including Thm 4.23) depends on it.\n\nWhat's actually new: the authors extend Kryliouk's construction of groups from simply-laced EALAs to arbitrary reduced types. They define cocycle-twisted Steinberg groups, build a Kac-Peterson integration group, and try to recover the extended affine Weyl group as Ad(N)/Ad(T). If true, that completes a natural gap in the literature. The organization is clear, and the paper is honest about relying on prior work: the Weyl group presentation from Azam, finite root system facts from [22], and the simply-laced base from Kryliouk's thesis.\n\nThe problem is real. Consequence 4.4 claims that if α+β is not a non-isotropic root, then (xα(t),xβ(s))=1, because R+_{α,β} is empty by Definition 3.7. But Definition 3.7 only defines nilpotent pairs, and there can be roots in R+_{α,β} even when α+β is not in R×. The stress-test example is the affine A1(1) case: take β=-α+δ with δ isotropic. Then α+β=δ is isotropic, but 2α+β=α+δ is a real root, so R+_{α,β} is not empty. In the affine Kac-Moody group, that commutator is not obviously 1; the St2 relation gives no information for this pair. The consequence is used in Prop 4.7 to make χ independent of the order of the product and to check it is a homomorphism. Without it, χ is not well-defined, and the epimorphism from St_{˙R}(Cσ) to St_{R,˙R,σ}(C) is not established. Proposition 4.12 and Theorem 4.23 both ride on that.\n\nOther spots: Prop 4.13(ii) has a 'one can see from classification' that is not spelled out and uniqueness is asserted without proof; Prop 4.22 is just 'similar to [16, Prop 3.2.41]' with no details; Theorem 4.23's relation (iii) is compressed. These would need work in any revision. The core issue is the Steinberg relation.\n\nVerdict: as written, the central theorem is unsupported. The authors need to repair the definition of the Steinberg group or supply a correct proof of the needed commutativity, and then rework Prop 4.7. This is worth sending to a knowledgeable referee—the goal is plausible and the community would care if fixed. But I wouldn't rely on it until the gap is addressed.\n\nRecommendation: engage as referee if asked, but expect heavy revision; do not cite as established.","headline":"A serious attempt to extend Kryliouk's simply-laced group construction to all reduced EALA types, but the central Steinberg-group map rests on a false commutativity claim, so the main theorem is unsupported as written.","tokens_in":18361,"tokens_out":6953,"would_cite":false,"duration_ms":129584,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["17B67","17B65","19C99","20G44","22E65"],"pacs":[],"model":"deepseek-v4-flash","headline":"For every tame extended affine Lie algebra with reduced root system, the extended affine Weyl group is the quotient of two natural subgroups of the associated integrated group.","keywords":["extended affine Lie algebras","extended affine root systems","Steinberg groups","Kac-Moody groups","Weyl groups","integration of Lie algebras","twisted group algebras"],"falsifier":"Take an affine Kac–Moody group (nullity one, trivial 2-cocycle) and compute the commutator of the one-parameter subgroups for two real roots whose sum is the imaginary root; a nontrivial commutator would contradict Consequence 4.4 and force a revision of the Steinberg group relations.","tokens_in":17262,"feed_emoji":"🧮","tokens_out":12940,"duration_ms":269484,"temperature":0.7,"pith_summary":"The paper builds group-theoretic companions for extended affine Lie algebras (EALAs), the higher-nullity generalizations of affine Kac–Moody algebras. It defines Steinberg groups attached to reduced extended affine root systems by generators and relations, then uses the integration method for integrable Lie algebras to associate a group to every tame EALA. Its main theorem states that the extended affine Weyl group of the algebra is isomorphic to the quotient $\\mathrm{Ad}(N)/\\mathrm{Ad}(T)$, where $\\mathrm{Ad}(N)$ and $\\mathrm{Ad}(T)$ are the images of the subgroups generated by the elements $n_\\alpha(t)$ and $h_\\alpha(t)$ in the adjoint group. This extends to arbitrary types a realization previously known only for simply laced root systems, placing the Weyl group of every tame EALA inside the structure of an associated group in the same way Kac–Moody groups encode Kac–Moody Weyl groups.","feed_headline":"Weyl groups of tame EALAs are quotients of two natural subgroups","feed_subtitle":"Steinberg groups attached to reduced extended affine root systems make the Weyl group a quotient, for every type.","key_machinery":"The load-bearing object is the Steinberg group $St_{R,\\dot R,\\sigma}(\\mathbb C)$ attached to a reduced extended affine root system $R$: generators $x_\\alpha(t)$ for non-isotropic roots $\\alpha$ and scalars $t$, with commutator relations taken from the finite rank-two subsystems $R_{\\alpha,\\beta}$, plus a rank-one relation when needed. Over it sits the Steinberg group $St_{\\dot R}(\\mathbb C_\\sigma)$ for the finite root system $\\dot R$ with coefficients in the twisted group algebra $\\mathbb C_\\sigma$, and a map $\\chi$ that sends $x_{\\dot\\alpha}(\\sum_\\delta t_\\delta c_\\delta)$ to the ordered product of the $x_{\\dot\\alpha+\\delta}(t_\\delta)$. The integration functor for integrable Lie algebras turns these generators into one-parameter subgroups of the integrated group, and the quotient $\\mathrm{Ad}(N)/\\mathrm{Ad}(T)$ is shown to have exactly the generalized presentation by conjugation of the extended affine Weyl group.","core_discovery":"The central claim is Theorem 4.23: for a tame extended affine Lie algebra $L$ over $\\mathbb C$ with a reduced extended affine root system $R$, the extended affine Weyl group $W$ is isomorphic to $\\mathrm{Ad}(N)/\\mathrm{Ad}(T)$, where $\\mathrm{Ad}(N)$ and $\\mathrm{Ad}(T)$ are the images in the adjoint group of the subgroups generated by the elements $n_\\alpha(t)$ and $h_\\alpha(t)$ for non-isotropic roots $\\alpha$ and $t\\in\\mathbb C^*$. The proof defines a Steinberg group $St_{R,\\dot R,\\sigma}(\\mathbb C)$ over the twisted group algebra $\\mathbb C_\\sigma$, maps it onto the integrated group of the algebra, and verifies that the images of the $n_\\alpha(t)$ satisfy the generalized presentation by conjugation of $W$. The isomorphism sends each reflection $w_\\alpha$ to the coset $\\mathrm{Ad}(T)\\mathrm{Ad}(n_\\alpha(t))$.","pith_inferences":["The construction isolates the commutativity of root subgroups with isotropic root sum as the key condition; if it holds, the same two-step template (Steinberg group over the twisted group algebra, then integration) could plausibly define extended affine Kac–Moody groups over fields other than $\\mathbb C$ or for non-reduced root systems.","The quotient $\\mathrm{Ad}(N)/\\mathrm{Ad}(T)$ may depend on the 2-cocycle $\\sigma$ through the finer structure of the integrated group; a natural test is whether different 2-cocycles give non-isomorphic groups with the same Weyl group quotient.","Because the proof only needs finite rank-two subsystems of the root system, analogues may exist for other classes of integrable Lie algebras whose root systems have finite rank-two slices, though this is not pursued in the paper."],"forward_implications":["Every tame extended affine Lie algebra with reduced root system carries a group whose two natural subgroups have as their quotient exactly the extended affine Weyl group.","For nullity-one affine Kac–Moody algebras and for finite root systems, the Steinberg groups constructed here coincide with the classical ones because the defining 2-cocycle is trivial.","The epimorphism from the Steinberg group to the integrated group gives explicit group elements realizing the generators of the Weyl group, including the 'reduced collection' relations beyond ordinary Coxeter relations.","The construction provides a uniform treatment for all reduced types, including cases where the extended affine Weyl group does not admit a Coxeter presentation."],"supporting_citations":[{"why":"Supplies the generalized presentation by conjugation of extended affine Weyl groups used in Theorem 4.23.","marker":"[7]"},{"why":"The simply-laced case this paper generalizes; its Steinberg and integration constructions are extended to arbitrary types.","marker":"[16]"},{"why":"Provides the commutator constants and rank-two root system facts used in the Steinberg relations.","marker":"[22]"},{"why":"The Kac–Moody generator-and-relations model that the extended affine construction follows.","marker":"[23]"},{"why":"Introduces the integration method for integrable Lie algebras used to attach a group to the EALA.","marker":"[15]"},{"why":"Supplies the SL2-subgroup and affine Kac–Moody facts used in the integration arguments.","marker":"[17]"},{"why":"The structural theory of EALAs and reduced extended affine root systems used throughout.","marker":"[1]"},{"why":"Gives the root-system and core facts used to identify the rank-two subalgebras generated by nilpotent pairs.","marker":"[4]"}],"fun_headline_variants":["Ad(N)/Ad(T) = Weyl group for tame EALAs","A quotient of two subgroups recovers the EALA Weyl group","Steinberg groups over twisted algebras give Weyl quotient","Tame EALA: Weyl group as a natural quotient","Weyl group of tame EALA from two natural subgroups"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction depends on the assertion that root subgroups commute whenever the sum of two non-isotropic roots is not itself a non-isotropic root; if that commutativity fails, the Steinberg group presentation and the epimorphism onto the integrated group are not established.","fun_headline_variants_meta":{"raw":{"variants":["Ad(N)/Ad(T) = Weyl group for tame EALAs","A quotient of two subgroups recovers the EALA Weyl group","Steinberg groups over twisted algebras give Weyl quotient","Tame EALA: Weyl group as a natural quotient","Weyl group of tame EALA from two natural subgroups"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001176,"raw_usage":{"total_tokens":4795,"prompt_tokens":814,"completion_tokens":3981,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":430,"completion_tokens_details":{"reasoning_tokens":3892}},"tokens_in":430,"tokens_out":3981,"duration_ms":31014,"temperature":1.0,"reasoning_tokens":3892,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:58:14.965203+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take an affine Kac–Moody group (nullity one, trivial 2-cocycle) and compute the commutator of the one-parameter subgroups for two real roots whose sum is the imaginary root; a nontrivial commutator would contradict Consequence 4.4 and force a revision of the Steinberg group relations.","supporting_citations":[{"cited_title":"Alg., 28 (2000), pp","cited_arxiv_id":null,"evidence_quote":"Supplies the generalized presentation by conjugation of extended affine Weyl groups used in Theorem 4.23."},{"cited_title":"Kryliouk, On the structure of quasi-simple Lie algebras and their auto morphism groups, 1995","cited_arxiv_id":null,"evidence_quote":"The simply-laced case this paper generalizes; its Steinberg and integration constructions are extended to arbitrary types."},{"cited_title":"Steinberg , Lectures on Chevalley Groups , Yale Uni., 1967","cited_arxiv_id":null,"evidence_quote":"Provides the commutator constants and rank-two root system facts used in the Steinberg relations."},{"cited_title":"Tits , Uniqueness and presentation of Kac-Moody groups over fields , J","cited_arxiv_id":null,"evidence_quote":"The Kac–Moody generator-and-relations model that the extended affine construction follows."},{"cited_title":"Kac , Constructing groups associated to infinite-dimensional Li e algebras , Inf","cited_arxiv_id":null,"evidence_quote":"Introduces the integration method for integrable Lie algebras used to attach a group to the EALA."},{"cited_title":"Moody and A","cited_arxiv_id":null,"evidence_quote":"Supplies the SL2-subgroup and affine Kac–Moody facts used in the integration arguments."},{"cited_title":"Allison, S","cited_arxiv_id":null,"evidence_quote":"The structural theory of EALAs and reduced extended affine root systems used throughout."},{"cited_title":"Allison and Y","cited_arxiv_id":null,"evidence_quote":"Gives the root-system and core facts used to identify the rank-two subalgebras generated by nilpotent pairs."}],"review_version":1}