{"id":"01b845e5-78c1-49bf-8ef6-c4ac60d79479","arxiv_id":"1908.04707","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Two-row shapes in affine type A admit finite W-graphs, constructed combinatorially and proven unique up to isomorphism, extending the known finite-type picture.","lead":"For affine symmetric groups, the paper constructs finite W-graphs for two-row Young diagram shapes by explicit combinatorial rules, and proves they are unique among admissible graphs with the same undirected part. These are the first non-trivial finite W-graphs in an affine type, giving explicit encodings of affine Hecke algebra representations.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central construction and uniqueness both rely on unverified 'verbatim' extensions of Stembridge, Chmutov, and Nguyen to non-bipartite admissible graphs; this unchecked dependence is the main risk.","rationale":"The reader's weakest assumption identifies exactly the point I would attack: the blanket assertion that dropping bipartiteness leaves the external theorems valid. My reading of the manuscript confirms that every central theorem (4.9, 8.6, 9.5, 9.10) either invokes Theorem 3.1 or one of the Section 7.2 transfers, and in each case the proof is a single sentence claiming the original proof carries through. Because the paper itself repeatedly flags this as the delicate point (Sections 1, 3.7, 7.2), it is not an artifact of the review process. I did not find an independent internal inconsistency; the concern is unsupported dependence, not a known contradiction. I also note secondary gaps in the Polygon Rule write-up (e.g., Section 6.2.3's 'same argument works verbatim' and Lemma 6.1's 'essentially verbatim'), but these are less load-bearing than the external transfer. The existing CONDITIONAL verdict is the right level; I would not raise or lower it without the audit in concrete_test.","tokens_in":52285,"tokens_out":9126,"duration_ms":92235,"concrete_test":"Independently audit [Ste08a, Section 4, especially Theorem 4.9], [Ngu18, Theorem 8.1], and [Chm15, Lemma 2.3.1] line by line, recording every use of 'bipartite' or of a parity/coloring argument. If any such use is needed for the compatibility, simplicity, or polygon implications, then Theorem 3.1 and Theorems 7.1-7.3 fail in the stated nb-admissible generality. As a secondary check, directly verify the Hecke algebra braid and commutation relations for the matrices of Γ_(3,2) (n=5) and Γ_(3,3) (n=6) over Z[q^{±1/2}]; a counterexample there would settle the concern, while a pass would isolate the risk to the proof audit.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim Theorem 4.9 is proved by checking Stembridge's four rules and invoking Theorem 3.1, which is asserted to extend [Ste08a, Theorem 4.9] from admissible to nb-admissible graphs because 'the original proof ... does not use the bipartition assumption' (Section 3.7). The same one-line transfer is used for Chmutov's Theorem 7.1, Nguyen's Theorems 7.2 and 7.3, and Chmutov's arc-transport Lemma 9.8. These extensions are genuinely load-bearing: Theorem 4.9 (existence of a finite affine W-graph) and Theorem 8.6 (uniqueness) collapse if any of them is false. Yet the paper supplies no demonstration, and the reader cannot check from the text alone whether bipartiteness is used in, e.g., defining vertex signs, proving the Simplicity Rule, or controlling the polygon equality. The internal Polygon Rule verification is also extremely long and has several 'same argument works verbatim' cases (Section 6.2.3), but the external-theorem transfer is the single most consequential gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a finite [1,n]-labeled graph Γ_λ on row-standard Young tableaux of a two-row partition λ, with moves of the first and second kind defined in Section 4.1, and claims in Theorem 4.9 that Γ_λ is a finite W-graph (S̃_n-graph) for the affine symmetric group. The proof checks Stembridge's four rules for nb-admissible graphs via a lengthy case analysis in Sections 5–6, using Theorem 3.1, an asserted extension of Stembridge's characterization to non-bipartite admissible graphs. The paper further proves uniqueness of such W-graphs for unequal two-row shapes (Theorem 8.6), minimality and maximality results for equal shapes (Theorems 9.6 and 9.10), and identifies the construction with a quotient of Lusztig's periodic W-graphs under a positivity assumption (Theorem 10.7).","tokens_in":52521,"tokens_out":5568,"duration_ms":49557,"significance":"If the main theorems hold, this is the first non-trivial family of finite W-graphs in an affine type with an explicit combinatorial rule, and it gives a strong uniqueness statement. The construction is concrete and testable: Γ_λ has no free parameters, its undirected part is the affine dual equivalence graph D_λ from [CLP17, CPY18], and the paper provides many worked examples. The proof structure is transparent, and the comparison with periodic W-graphs suggests a general conjecture. The main weakness is the repeated reliance on unproved 'verbatim' extensions of external theorems to the non-bipartite setting, which are load-bearing for both existence and uniqueness.","major_comments":[{"comment":"The proof of Theorem 3.1 is a one-sentence appeal to [Ste08a, Theorem 4.9] with the assertion that Stembridge's proof 'does not use the bipartition assumption.' This is not a proof. The notion of nb-admissible graphs is new to this paper, and the bipartiteness condition may be used implicitly in Stembridge's arguments (for example, in defining the sign conventions in the Simplicity Rule or in the derivation of the Polygon Rule). Since Theorem 4.9 invokes Theorem 3.1 directly, the existence of Γ_λ as a W-graph is not independently verifiable from the manuscript. Please provide a complete proof of Theorem 3.1, or at minimum a detailed line-by-line verification that each step of Stembridge's proof survives without bipartiteness, with specific references to [Ste08a].","section":"§3.7, Theorem 3.1"},{"comment":"The same verbatim-transfer argument is used for Chmutov's Theorem 7.1, Nguyen's Theorems 7.2 and 7.3, and Chmutov's arc-transport Lemma 9.8. These results are not peripheral: Theorem 7.1 underlies the description of cells in Proposition 7.5, Theorem 7.2 and Theorem 7.3 are used in the proof of uniqueness (Theorem 8.6) and in Theorems 9.6 and 9.10, and Lemma 9.8 is used in the maximality argument. The paper does not identify where bipartiteness is used (or not used) in the original sources. As a result, the uniqueness theorem inherits the same gap. Please provide either full proofs adapted to nb-admissible graphs or precise statements of the corresponding theorems in the original setting and a reduction from the nb-admissible case to the admissible case.","section":"§7.2, Theorems 7.1–7.3 and Lemma 9.8"},{"comment":"The Polygon Rule verification for the four-element case (Section 6) is the core technical argument for Theorem 4.9, but several subcases are not actually written out. In Section 6.2.3, the cases d = a+2, d = a−1, b = a−1, b = a+2 are each dismissed with 'the same argument works verbatim' from generic cases, and in Lemma 6.1 the case {i,j} = {a,b} is declared 'essentially verbatim' after a substitution. The reader cannot check whether the inequalities in conditions (d) and (e) of Definition 4.1 behave identically in these special cases (for instance, when intervals such as ⌜d+2, a−2⌟ become empty). Please write out these subcases or give a precise reduction that proves the equality from the generic case.","section":"§6.2.3 and Lemma 6.1"}],"minor_comments":[{"comment":"In the paragraph following the definition of the set T, the expression 'n − 2k ∈ f T' appears to be a typo for 'n − 2k ∈ T'.","section":"§5.3.2"},{"comment":"The condition 'i ⁄∈ {j − 1, j, j + 1}' is typeset with a nonstandard symbol; it should be the standard 'i ∉ {j−1, j, j+1}'.","section":"§4.3"},{"comment":"The term 'dual Knuth move' is used without a definition in the text; the reader must infer it from the context of [Hai92].","section":"§7.3, Proposition 7.4"},{"comment":"The references [Fun03a] and [Fun03b] appear to refer to the same paper; the text uses both labels in a way that may confuse readers.","section":"References"},{"comment":"The construction of Γ^quot_λ involves finite sums over infinite γ-orbits; the paper states that finiteness follows from [Var04] and [Lus97, Consequence 13.8], but the precise statement of that consequence is not quoted, making the finiteness claim hard to verify without consulting the sources.","section":"§10.5"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically ambitious and likely important if the gaps are filled. The most effective revision would be a supplement containing the missing 'verbatim' proofs, since the current text places a heavy verification burden on the reader. I do not see grounds for rejection; the issues are localizable and fixable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The construction is the real thing: for two-row shapes in affine type A, the paper gives explicit move rules, proves they form a finite W-graph, proves uniqueness for unequal rows, and compares the result with quotients of Luszttig's periodic W-graphs under a stated positivity assumption. That is the first nontrivial family of finite W-graphs in an affine type, and it answers a question left open after the affine dual equivalence work of Chmutov, Pylyavskyy, Yudovina, and Lewis. The authors deserve credit for an explicit, example-rich construction and for flagging the conditional nature of the Section 10 comparison rather than overselling it.\n\nThe soft spot is exactly where the stress-test lands. Theorem 3.1 asserts that Stembridge's characterization of W-graphs remains valid when bipartiteness is dropped from admissibility, and Theorem 4.9 depends on it. The same one-line transfer is used for Chmutov's theorem, Nguyen's theorems, and the arc-transport lemma. The paper says the original proofs do not use the bipartition assumption and carry through essentially verbatim, but it does not show this. Since bipartiteness could matter for vertex signs or for parts of the polygon argument, and since the existence and uniqueness theorems both rest on this transfer, this is a load-bearing gap. It is not obviously fatal — the claim may well be true — but the reader cannot verify it from the text without going back to the sources and rechecking every use of the bipartition assumption.\n\nThe Polygon Rule verification is also long and partly case-by-case, with several subcases dispatched as 'the same argument works verbatim.' That is harder to referee, but it is detailed, and the pattern of the argument looks coherent rather than hand-wavy. The finite-type restriction in Section 7.4 is a repackaging of a known Lascoux-Schutzenberger description, which is fine and clearly credited.\n\nIf the nb-admissible transfer is wrong, both the central existence theorem and the uniqueness theorems collapse. That justifies a conditional verdict, not a rejection. The paper is honest, technically serious, and likely correct in its main ideas. Send it to a good referee with instructions to check the Stembridge/Chmutov/Nguyen extensions line by line; the authors should be asked to provide actual proofs or a supplement. The field needs these examples, and with that gap closed the paper would be a solid contribution.","headline":"First family of finite affine-type W-graphs, with a real and potentially repairable gap: the paper repeatedly extends three external theorems to non-bipartite graphs without proof.","tokens_in":53121,"tokens_out":1768,"would_cite":true,"duration_ms":21600,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05E10","20C08"],"pacs":[],"model":"deepseek-v4-flash","headline":"For two-row partitions of n, this paper constructs finite W-graphs for the affine symmetric group, proves their uniqueness under admissibility, and identifies them with quotients of periodic W-graphs under a positivity assumption.","keywords":["W-graphs","affine symmetric group","two-row partitions","Young tableaux","Hecke algebra","admissible graphs","affine dual equivalence","periodic W-graphs"],"falsifier":"Check whether the standard admissibility characterization really survives without bipartiteness: either the original proof uses bipartiteness at some step, or one can find a non-bipartite nb-admissible graph satisfying the four rules that is not a W-graph. A computational search over small finite graphs with descent-like labels would settle the theorem that certifies Γ_λ.","tokens_in":52053,"feed_emoji":"🕸️","tokens_out":6965,"duration_ms":64075,"temperature":0.7,"pith_summary":"The paper constructs, for every two-row partition of n, a finite directed graph Γ_λ whose vertices are row-standard Young tableaux of shape λ and whose edges are given by two explicit combinatorial moves. It proves that Γ_λ is a W-graph for the affine symmetric group — the first non-trivial infinite family of finite W-graphs in an affine Coxeter type — so it encodes a finite-dimensional module for the affine Hecke algebra. The undirected edges of Γ_λ recover the known affine dual equivalence graph, and the paper proves that any W-graph with the same undirected skeleton and satisfying a mild admissibility condition is isomorphic to Γ_λ when the two rows have different lengths. In the equal-length case the construction is almost unique, with explicit minimal and maximal variants. The paper further shows that, if the coefficients of the periodic W-graph construction are nonnegative, Γ_λ is isomorphic to the corresponding quotient of that periodic graph.","feed_headline":"Two-row shapes yield first finite affine W-graphs","feed_subtitle":"Explicit tableaux moves build W-graphs for the affine symmetric group, then prove uniqueness and a link to periodic graphs.","key_machinery":"The load-bearing object is Γ_λ: the [1,n]-labeled graph whose vertices are row-standard Young tableaux of shape λ, whose labels are descent sets, and whose directed edges are the first-kind and second-kind moves, where second-kind moves are governed by five explicit arithmetic conditions involving cyclic intervals. This graph is checked against four Stembridge rules for admissible W-graphs; the polygon rule, requiring equality of path counts N^r_{ij}(u,v)=N^r_{ji}(u,v) over two- and three-step paths, is the technical heart and is proven by a detailed case analysis on pairs of tableaux differing in two or four entries. The same graph and its restriction to the finite symmetric group then serve as the uniqueness classifier: the admissibility theorems for type-A W-graphs, taken in their non-bipartite form, identify cells with standard graphs and force the directed edges of any competing graph to match those of Γ_λ.","core_discovery":"Γ_λ is an [1,n]-labeled graph on RSYT(λ) with descent labels and edge weights 0 or 1. Its directed edges are of two kinds: first-kind moves swap i and i+1 between the two rows, while second-kind moves swap i from the second row with j from the first row subject to five explicit conditions about odd differences, neighboring entries, and interval counts. The main theorem is that Γ_λ satisfies the four local rules — compatibility, simplicity, bonding, and polygon — and therefore, by the extended admissibility theorem, is a true W-graph. The polygon rule, which equates two-step path counts, is verified case by case according to whether the endpoints differ by two or by four entries. The undirected skeleton U(Γ_λ) is exactly the affine dual equivalence graph D_λ, and the restriction of Γ_λ to the finite symmetric group splits into cells isomorphic to standard type-A W-graphs, labeled by semistandard tableaux. The uniqueness theorem states that when λ=(λ1,λ2) with λ1>λ2, any nb-admissible W-graph with underlying graph D_λ is isomorphic to Γ_λ; in the equal-row case the paper exhibits a minimal graph Γ'_λ embedded in every such graph and shows that Γ_λ is maximal among translation-invariant ones. Under the conjectural nonnegativity of periodic W-graph coefficients, Γ_λ is isomorphic to the quotient Γ^quot_λ of the periodic W-graph.","pith_inferences":["One implicit consequence is that the bipartiteness hypothesis may be safely dropped throughout the theory of admissible W-graphs; if so, the same four-rule characterization can be used to search for finite W-graphs in other affine types.","The move rules for Γ_λ resemble an affine analogue of Knuth moves, so they may yield a new notion of affine dual equivalence useful for other representation-theoretic statistics.","The equal-length case hints at a whole family of intermediate graphs parameterized by integer weights on cross-component edges; the paper notes this possibility but does not develop it, so one could test systematically which of those intermediate graphs are W-graphs.","The isomorphism with quotients of periodic W-graphs is conditional on a positivity conjecture; if that conjecture holds, the explicit moves give a combinatorial handle on canonical bases in affine Springer theory."],"forward_implications":["If correct, the paper supplies the first infinite family of finite W-graphs in an affine type that is built purely combinatorially, without representation-theoretic input.","Every two-row affine dual equivalence graph can be enriched with directed edges to a genuine W-graph, resolving the enrichment question for these shapes.","When the two rows have unequal length, the W-graph is pinned down by its undirected skeleton: any nb-admissible W-graph with that skeleton is the same graph, so the directed edges are not extra free data.","Restricting Γ_λ to the finite symmetric group gives modules whose Frobenius characters are Hall-Littlewood symmetric functions, yielding a tableau model for those representations.","If the periodic-W-graph coefficients are nonnegative, the affine two-row W-graphs coincide with quotients of periodic W-graphs, connecting the construction to canonical bases and Springer theory."],"supporting_citations":[{"why":"Provides the four-rule characterization of admissible W-graphs that the paper extends by dropping bipartiteness and then uses to certify Γ_λ.","marker":"[Ste08a]"},{"why":"Shows that in type A each cell of an admissible W-graph is a single simple component with a dual equivalence graph as skeleton, a fact used in the restriction and uniqueness arguments.","marker":"[Chm15]"},{"why":"Proves that admissible type-A cells are exactly the standard type-A W-graphs, which the paper invokes to force directed edges in the uniqueness theorems.","marker":"[Ngu18]"},{"why":"Defines the affine dual equivalence graph D_λ and describes the molecules that form the undirected skeleton matched by Γ_λ.","marker":"[CLP17]"},{"why":"Introduces affine dual equivalence graphs as finite quotients and provides the affine matrix-ball construction used in the comparison with periodic W-graphs.","marker":"[CPY18]"},{"why":"Defines periodic W-graphs whose quotients are compared to Γ_λ under the nonnegativity assumption.","marker":"[Lus97]"},{"why":"Proves that periodic W-graph coefficients in type A are finite, making the quotient construction well-defined.","marker":"[Var04]"},{"why":"Originates W-graphs and cell representations of Hecke algebras, the object class being constructed and extended here.","marker":"[KL79]"}],"fun_headline_variants":["First finite affine W-graphs via two-row shapes","Two-row tableaux give unique finite W-graphs in affine type A","Two-row W-graphs: finite, unique, first in affine type","Two-row shapes build finite affine W-graphs, uniquely"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the assertion that the standard structure theorems for admissible W-graphs remain true when the bipartiteness condition is removed, with proofs said to carry over essentially verbatim even though no demonstration is supplied.","fun_headline_variants_meta":{"raw":{"variants":["First finite affine W-graphs via two-row shapes","Two-row tableaux give unique finite W-graphs in affine type A","Two-row W-graphs: finite, unique, first in affine type","Two-row shapes build finite affine W-graphs, uniquely"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00116,"raw_usage":{"total_tokens":4783,"prompt_tokens":905,"completion_tokens":3878,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":521,"completion_tokens_details":{"reasoning_tokens":3802}},"tokens_in":521,"tokens_out":3878,"duration_ms":28145,"temperature":1.0,"reasoning_tokens":3802,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:35:19.350868+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check whether the standard admissibility characterization really survives without bipartiteness: either the original proof uses bipartiteness at some step, or one can find a non-bipartite nb-admissible graph satisfying the four rules that is not a W-graph. A computational search over small finite graphs with descent-like labels would settle the theorem that certifies Γ_λ.","supporting_citations":[],"review_version":1}