{"id":"95a2f273-c412-4917-b8ee-0370761de244","arxiv_id":"2507.02347","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A categorical embedding Ψ for extended affine type A Soergel bimodules is constructed, categorifying the Hecke algebra parabolic embedding, with an example categorifying the Zelevinsky tensor product of two trivial representations.","lead":"The authors construct a monoidal functor that categorifies the parabolic embedding of extended affine type A Hecke algebras, the first step toward categorifying the Zelevinsky tensor product. A worked example categorifies the two-dimensional representation W of the n=2 Hecke algebra, with a conjectural isomorphism of triangulated Grothendieck groups.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.9 rests on unverified naturality computations; a failure in Lemma 4.12(106) or the N8 direct check would invalidate the symmetric pair condition.","rationale":"The reader's verdict accepts with moderate confidence, and our stress-test agrees with the reader's weakest assumption. The central construction Ψ_{k,n−k} is a monoidal functor only if Ψ_L and Ψ_R are well-defined and form a symmetric pair. The well-definedness of Ψ_L is proven in detail, but the proof for Ψ_R is skipped as 'very similar', which is a missing support though likely fillable. The symmetric pair condition, however, depends on the naturality of the complicated braiding ζ, whose proof relies on the ζ-box relations in Lemma 4.12 and on direct computations in N7/N8 that are not machine-checked. The two exceptional cases in Lemma 4.11 (morphism spaces of dimension 3 and 5) are handled by direct diagrammatic computations that are notoriously error-prone. If any of these identities fails, Theorem 4.9 would fail. This is not an internal inconsistency, but a correctness risk due to computational burden. The paper is honest about its conjectures and limitations (e.g., Conjecture 4.15, Conjecture 5.29). It provides extensive examples and detailed proofs, and the field-standard diagrammatic method is acceptable. Therefore the verdict remains ACCEPT, but the proposed concrete test would substantially de-risk the central claim.","tokens_in":58157,"tokens_out":17511,"duration_ms":190359,"concrete_test":"For n = 4, k = 2, write both sides of Lemma 4.12(106) and the N8 naturality square (113) explicitly as morphisms of bounded chain complexes of Soergel bimodules over R, using the algebraic definitions in [MMV, §4.1], and check that they are chain homotopic by constructing the homotopy or using a computer algebra system for Soergel bimodules. If either equality fails, Lemma 4.8(b) and hence Theorem 4.9 would be false.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim Theorem 4.9 depends on Lemma 4.8(b), which asserts that Ψ_L and Ψ_R form a symmetric pair. The naturality of ζ, proved in (N1)–(N12), is the least robust part: most cases use the hom & dot trick, which requires the relevant morphism space to be one-dimensional and the dotting operation to be injective. Lemma 4.11 explicitly identifies two exceptions (X = B_0 with a dot or a trivalent vertex labeled 0) where the morphism space has dimension 3 or 5; these are handled by direct diagrammatic computations (N7, N8). The N8 computation is a long sequence of diagram rewrites, and the ζ-box relations in Lemma 4.12(105)–(106) are proved by equally long computations using relations (80), (68), (45), (61), (57), (59), and (77). A single misapplied relation in any of these steps would break naturality of ζ, hence the symmetric pair condition, and with it the monoidal functor Ψ_{k,n−k}. Additionally, the paper proves well-definedness only for Ψ_L, stating that Ψ_R is 'very similar', which is a missing support, though likely fillable. These computations are not machine-checked and no independent algebraic verification is offered.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a monoidal functor Ψ_{k,n−k} : bSext_k ⊠ bSext_{n−k} → K^b(bSext_n) for 1 ≤ k < n, categorifying the parabolic embedding ψ_{k,n−k} of extended affine type A Hecke algebras. The construction is carried out through a pair of monoidal functors Ψ_L, Ψ_R and a symmetric-pair braiding ζ, with the main theorem (Theorem 4.9) reducing to well-definedness and naturality checks in the diagrammatic Soergel calculus. The paper also develops new diagrammatic relations for Rouquier complexes in extended affine type A, proves the main theorem via those relations, and then works out an explicit example: the categorified Zelevinsky tensor product of the trivial birepresentation of bSext_1 with itself. The example leads to a wide finitary cover U, a triangulated orbit category W, and a conjecture (Conjecture 5.29) relating the triangulated Grothendieck group of W to the decategorified representation W of bH_ext_2.","tokens_in":58399,"tokens_out":9356,"duration_ms":102907,"significance":"If the main theorem is correct, this is a significant step toward the categorification of the Zelevinsky tensor product for extended affine type A Hecke algebras, a setting where earlier categorical induction functors for finite type A do not directly apply. The paper is careful to benchmark the construction against the Soergel categorification theorem and Soergel's hom formula, and it clearly separates proved results from conjectures (Conjectures 4.15 and 5.29). The explicit example in Section 5.3, including the construction of the wide finitary cover U and the triangulated orbit category W, is a useful test of the framework. The main weakness is the heavy computational verification burden: many essential diagrammatic identities are summarized rather than fully demonstrated, and the computations are not machine-checked. This makes the proof of Theorem 4.9 convincing only to the extent that the reader is willing to trust the sketched 'hom & dot trick' arguments and the assertion that the Ψ_R check is 'very similar'.","major_comments":[{"comment":"The well-definedness of the functors is proved only for Ψ_L. The text states: 'The proofs that Ψ_L and Ψ_R preserve all diagrammatic relations are very similar, so we only prove that Ψ_L is well-defined.' Since Lemma 4.8(a) and Theorem 4.9 require both functors, and since Ψ_R is defined by genuinely different formulas on the generators (for example Ψ_R(B_0) = T_0 ... T_{k−1} B_k T^{-1}_{k−1} ... T^{-1}_0 and Ψ_R(B_ρ) = T^{-1}_k ... T^{-1}_1 B_ρ), it is not immediate that the same verification applies, in particular for the relations (34)–(37) and (46) that involve the labels 0 and k−1. The authors should either provide the analogous proof for Ψ_R or exhibit a formal symmetry of the diagrammatic calculus that reduces Ψ_R to Ψ_L.","section":"§4.2.1, Lemma 4.8(a)"},{"comment":"The naturality of ζ is the core of the symmetric pair condition in Lemma 4.8(b), and hence of the main theorem, but the proof is not written out in full. Many cases (N3, N4, N5, N9, N10, N11, N12) are summarized with phrases such as 'we use the hom & dot trick' and 'the resulting diagram, in both cases, is ...', without explicitly verifying the two hypotheses of the trick in each case: that the relevant morphism space is one-dimensional and that the dotting operation is injective. Lemma 4.11 identifies the exceptions that need direct checks, but for the remaining cases the verification is left to the reader. For example, in (N3) the equality (108) is asserted after attaching a dot 'to the top endpoints labeled k+1', but no computation is shown to establish that the two resulting diagrams coincide. Since a single misapplied relation in any of these steps would invalidate Theorem 4.9, I ask for a systematic presentation of these checks, or a computer-verified appendix, so that the proof can be independently audited.","section":"§4.2.3, naturality of ζ"},{"comment":"The recursive definition of ζ_{X,Y} on arbitrary tensor products via formula (99) is said to satisfy the hexagon identities 'by definition'. The interchange law (95) is used to show that the two expansions (99) and (100) agree, but the well-definedness of the recursion with respect to different bracketing of three or more tensor factors (for example, ((X_1 X_2) X_3)Y versus (X_1 (X_2 X_3))Y) is not demonstrated. In a strict monoidal category the associators are identities, but the composite isomorphisms built from the ζ_{X_i,Y_j} could still depend on the order in which the hexagon identities are applied. The authors should state explicitly that the hexagon identities are being imposed as part of the recursive definition, explain why the resulting assignment is coherent, or cite a standard coherence theorem for braidings that covers this situation.","section":"§4.2.2, definition of ζ"}],"minor_comments":[{"comment":"The sentence 'because they m for 1≤m≤n all commute with each other' appears to contain a typo; it should presumably read 'because the y_m for 1≤m≤n all commute with each other'.","section":"§2.2"},{"comment":"The phrase 'The cases X=B_0, g= 1/0 and and X=B_ρ, g= n−k−1/0' contains a duplicated 'and'; the second 'and' should be removed.","section":"§4.2.3, case (N12)"},{"comment":"In the displayed computation for the k=2 case of relation (37), the expression 'nX j=2 j 1' would be easier to follow if the summation index and the labels on the strands were clarified; currently the notation is ambiguous because the summands depend on the color j.","section":"§4.2.1, k=2 case"},{"comment":"The definition of the orbit category bΩ uses morphisms A → B Y_{r,s}, and composition is defined as in (136). Since the Z^2-action is strong but not strict, the paper should spell out the associativity and unit constraints for the action, because the composition rule in (136) depends on the specific choice of the isomorphisms Y_{r,s}Y_{r',s'} ≅ Y_{r+r',s+s'}.","section":"§5.3.2, orbit category"},{"comment":"A table summarizing the notation for T_{[a,b]}, the interval diagrams, and the various ζ-box diagrams would improve readability. Many diagrams rely on labels that are only described in the surrounding text, and the reader must frequently switch between the displayed diagrams and the prose to determine the missing labels.","section":"§3.2 and §4.2.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely correct, and the main theorem is an important contribution, but my confidence is moderated by the computational burden of the proof. The missing proof of well-definedness for Ψ_R is a clear gap in a load-bearing lemma. The naturality checks for ζ are summarized to the point that an independent referee cannot easily verify them; I would be satisfied if the authors provided a supplementary appendix with the full computations or a computer-checkable verification of the diagrammatic relations (in particular Lemma 4.12 and cases N1–N12). If they do so, I would be happy to accept."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper does what its title says: it takes the first real step toward categorifying the Zelevinsky tensor product for extended affine type A. The new construction is the monoidal functor Psi_{k,n-k} and, more specifically, the symmetric pair (Psi_L, Psi_R) with its braiding zeta. The authors are upfront that the notion of a symmetric pair of monoidal functors seems to have no reference in the literature, and they give a clean abstract justification via Lemma 4.4. That part reads well.\n\nWhat the paper does best is the concrete example in Section 5. The decategorified story is worked out in enough detail to understand what is being categorified, and the constructions of the wide finitary cover U and the triangulated orbit category W are explicit. Conjecture 5.29 is clearly labeled as a conjecture, and the paper does not overclaim: it says where the missing steps are, including the unproved faithfulness conjecture and the non-trivial extension to K^b. That honesty is a real strength.\n\nThe soft spots are exactly where the stress-test note points. Theorem 4.9 rests on Lemma 4.8(b), and the naturality of zeta is proved by the hom & dot trick in most cases, with two exceptional cases (Lemma 4.11) checked by direct diagrammatic computations. The N8 computation is long, and a single misapplied relation would break the symmetric pair condition. This is not a fatal flaw in my reading: the proof strategy is systematic, the one-dimensionality checks are plausible, and the paper states its conventions carefully. But it is a load-bearing computational burden, and it is not machine-checked. The paper also proves well-definedness only for Psi_L, saying Psi_R is very similar; that is a missing detail, though it looks fillable.\n\nI checked the circularity worry and I do not see a real problem. The central claim is a new functor, not a reduction to earlier results. Self-citations to [MMV] and [MaTh] are used as background, and the hom formula is an external benchmark, not a fitted parameter. The citation pattern is normal for an ongoing program.\n\nWho will get value from this? People working on categorical representation theory of Hecke algebras and on birepresentations of wide finitary categories, especially those already following the MMV/MaTh program. For a general representation theorist it is probably too technical, but for the subfield it is a meaningful, honest contribution.\n\nMy recommendation: send it to peer review. It deserves referee time. Ask the referee to check the N8 computation and the Psi_R well-definedness claim carefully, and consider asking for a supplementary file or an appendix with the full computation. If those checks pass, accept.","headline":"A solid, carefully honest first step toward categorifying Zelevinsky tensor products; the central functor is plausible and the proof style is standard, but the main theorem leans on long diagrammatic computations that deserve close scrutiny.","tokens_in":808,"tokens_out":759,"would_cite":true,"duration_ms":34264,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["20C08","18M05","18G80"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper constructs a monoidal functor from the box tensor product of extended affine type A Soergel categories into the bounded homotopy category, categorifying the parabolic embedding that defines Zelevinsky tensor products.","keywords":["extended affine Hecke algebras","Soergel bimodules","Rouquier complexes","categorification","Zelevinsky tensor product","parabolic induction","birepresentations","triangulated orbit categories"],"falsifier":"Compute both sides of the equality in Lemma 3.8 for a concrete small instance, say $n=5$, $a=1$, $b=2$, $c=4$, inside $K^b(\\mathit{bSext}_5)$; if the two diagrams differ by a non-null-homotopic morphism, then that relation, and with it Theorem 4.9, fails.","tokens_in":57946,"feed_emoji":"🧶","tokens_out":12628,"duration_ms":134125,"temperature":0.7,"pith_summary":"The paper's goal is to lift parabolic induction for extended affine type A Hecke algebras to the categorical level. For each $0<k<n$ there is an algebra embedding $\\psi_{k,n-k}\\colon \\mathit{bH}^{\\mathrm{ext}}_k \\otimes \\mathit{bH}^{\\mathrm{ext}}_{n-k} \\to \\mathit{bH}^{\\mathrm{ext}}_n$ that underlies the Zelevinsky tensor product of finite-dimensional representations. The paper defines an $R$-linear monoidal functor $\\Psi_{k,n-k}\\colon \\mathit{bSext}_k \\boxtimes \\mathit{bSext}_{n-k} \\to K^b(\\mathit{bSext}_n)$ whose decategorification is $\\psi_{k,n-k}$, built from a symmetric pair of diagrammatic functors and a braiding $\\zeta$. If correct, this is a first step toward a categorified Zelevinsky tensor product, where induction of representations would happen at the level of Soergel bimodules and complexes.","feed_headline":"Parabolic induction in extended affine type A is categorified","feed_subtitle":"A monoidal functor lifts the Hecke embedding to Soergel bimodule complexes, opening the way to categorified induction.","key_machinery":"The central mechanism is a symmetric pair of monoidal functors: two strict monoidal functors $\\Psi_L$ and $\\Psi_R$ from the diagrammatic Bott-Samelson categories $\\mathit{cBS}^{\\mathrm{ext}}_k$ and $\\mathit{cBS}^{\\mathrm{ext}}_{n-k}$ into $K^b(\\mathit{bSext}_n)$ whose composites commute up to a natural isomorphism $\\zeta$ satisfying the hexagon identities. Lemma 4.4 turns such a pair into one monoidal functor on the box tensor product, and the explicit $\\zeta$ constructed in Section 4.2.2 makes $\\Psi_{k,n-k}$ monoidal. The technical engine is the Rouquier-Soergel diagrammatic calculus: the new relations in Lemmas 3.6, 3.8, 3.10, 3.11 and 4.12 are proved by the hom-and-dot trick, using Soergel's hom formula to reduce equalities to one-dimensional morphism spaces.","core_discovery":"The paper's central claim is Theorem 4.9: for $0<k<n$, the symmetric pair $\\Psi_L,\\Psi_R$ gives an $R$-linear monoidal functor $\\Psi_{k,n-k}\\colon \\mathit{bSext}_k \\boxtimes \\mathit{bSext}_{n-k} \\to K^b(\\mathit{bSext}_n)$ that categorifies the algebra embedding $\\psi_{k,n-k}$. Lemma 4.8 splits the work into well-definedness of the two diagrammatic functors and the existence of a natural braiding $\\zeta$ satisfying the hexagon identities; both parts are proved by checking diagram relations. The checks use new relations among Rouquier complexes and the hom-and-dot trick, which compares diagrams in one-dimensional morphism spaces via Soergel's hom formula. The paper also works out the first categorical induction example: the algebra object $Y=\\Psi_{1,1}(X\\boxtimes X)$ yields a wide finitary birepresentation $\\mathsf{U}$ of $\\mathit{bSext}_2$ decategorifying the infinite-dimensional module $U$, and a dg-enhanced orbit category $\\mathcal{W}$ is conjectured to categorify the induced representation $W=V\\odot V$.","pith_inferences":["If Theorem 4.9 is correct, the same push-forward of algebra objects through $\\Psi_{k,n-k}$ should yield induced triangulated birepresentations for arbitrary finite-dimensional modules, not only for the example $V\\odot V$.","The symmetric-pair technique may extend to parabolic embeddings of other types or to longer compositions $k_1+\\cdots+k_m=n$, as long as the corresponding $\\zeta$-box relations can be proved.","Because the new diagram relations are finite and explicit, machine-checked verification of Lemmas 3.6, 3.8, 3.10, 3.11 and 4.12 would be a natural and feasible next step.","The dg-enhanced orbit-category construction used for $\\mathcal{W}$ could serve as a general template for triangulated birepresentations in wide finitary settings, where a balanced box tensor product is not yet available."],"forward_implications":["The functor $\\Psi_{k,n-k}$ decategorifies to $\\psi_{k,n-k}$, so the diagrammatic construction is a genuine categorification of parabolic induction for extended affine type A.","Algebra objects in $\\mathit{bSext}_k \\boxtimes \\mathit{bSext}_{n-k}$ can be pushed forward to algebra objects in $K^b(\\mathit{bSext}_n)$, giving a mechanism to induce birepresentations; the paper's worked example is $W=V\\odot V$.","The wide finitary birepresentation $\\mathsf{U}$ categorifies the infinite-dimensional module $U$, and Theorem 5.22 determines its indecomposable objects up to isomorphism.","The triangulated orbit category $\\mathcal{W}$ is a $\\mathit{bSext}_2$-birepresentation, and Conjecture 5.29 predicts its triangulated Grothendieck group is the simple induced module $W_{\\mathbb{C}(q)}$.","The paper leaves two extensions for future work: an iterated functor $\\Psi_{k_1,\\dots,k_m}$ categorifying multi-parameter embeddings, and a natural isomorphism between the two-step composite embeddings."],"supporting_citations":[{"why":"Supplies the diagrammatic calculus for Rouquier complexes and the homotopy equivalences that the new relations and the definitions of $\\Psi_L$ and $\\Psi_R$ build on.","marker":"[MMV]"},{"why":"Establishes the extended affine Soergel categorification theorem, giving $\\mathit{bSext}_n$ and the Kazhdan-Lusztig basis that identifies the decategorification.","marker":"[MaTh]"},{"why":"Provides the Soergel hom formula, localization, and standard Soergel bimodule facts used in dimension and indecomposability arguments.","marker":"[EMTW]"},{"why":"Gives the extended affine Hecke presentation and the quotient category $\\mathcal{S}^{\\mathrm{ext}}_2$ needed for the worked example.","marker":"[Eli2]"},{"why":"Defines the Zelevinsky tensor product and the multisegment classification that the categorification ultimately targets.","marker":"[Zel]"},{"why":"Supplies the irreducibility criterion for Zelevinsky tensor products used to identify the simple module $W$.","marker":"[LNT]"},{"why":"Provides the wide finitary birepresentation framework and the correspondence with algebra objects used for the categorical induction.","marker":"[Macph]"},{"why":"Constructs dg-enhanced triangulated orbit categories, used to endow $\\mathcal{W}$ with a triangulated birepresentation structure.","marker":"[FKQ]"}],"fun_headline_variants":["Extended affine type A induction categorified","Soergel bimodule functor lifts induction in type A","Categorified induction in extended affine type A","Monoidal functor categorifies Hecke induction in type A"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The main theorem stands on the correctness of the new diagram relations for Rouquier complexes in extended affine type A; these relations are checked by hand in one-dimensional morphism spaces, so an error in any single relation would invalidate the functor.","fun_headline_variants_meta":{"raw":{"variants":["Extended affine type A induction categorified","Soergel bimodule functor lifts induction in type A","Categorified induction in extended affine type A","Monoidal functor categorifies Hecke induction in type A"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000511,"raw_usage":{"total_tokens":2420,"prompt_tokens":811,"completion_tokens":1609,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":427,"completion_tokens_details":{"reasoning_tokens":1545}},"tokens_in":427,"tokens_out":1609,"duration_ms":13348,"temperature":1.0,"reasoning_tokens":1545,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:31:04.779401+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute both sides of the equality in Lemma 3.8 for a concrete small instance, say $n=5$, $a=1$, $b=2$, $c=4$, inside $K^b(\\mathit{bSext}_5)$; if the two diagrams differ by a non-null-homotopic morphism, then that relation, and with it Theorem 4.9, fails.","supporting_citations":[],"review_version":1}