{"id":"8af228b3-cdc2-4fab-9b5f-1122c4233c11","arxiv_id":"2412.02281","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The authors compute explicit q-Stokes matrices for a confluent hypergeometric q-difference system and prove that the q-to-1 limit recovers classical Stokes matrices in a selected sector.","lead":"This paper derives an explicit connection formula for the basic hypergeometric series n-phi-(n-1) with one zero numerator parameter, then uses it to compute q-Stokes matrices for a confluent hypergeometric q-difference system with an irregular singularity. The result matters because it gives concrete Stokes data for a family of q-difference systems and shows the formulas reduce to known differential Stokes matrices as q tends to 1.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.1/3.12's displayed b_q formula is wrong as printed: the minor Δ_{1..\\hat{k}..n-1}^{1..n-2} appears in the denominator, while the proof's equation (50) and block conjugation place it in the numerator.","rationale":"The reader's weakest-assumption concern focused on the uniqueness step in Theorem 2.26's proof (identifying ψ(z) with h_q(z)_n). That is a proof-gap concern, but the identification is standard and essentially supported by Proposition 2.5's unique recursive determination of the formal series. My stress-test found a more concrete and load-bearing problem: the explicit b_q formula in the main theorem (Theorem 1.1/3.12) is misstated. The derivation in equation (50), together with Proposition 3.4, shows that the minor Δ_{1..\\hat{k}..n-1}^{1..n-2}(A-λ_j I) should multiply the numerator, not divide it. For generic matrices this changes the numerical value of every entry of b_q, so the theorem as printed gives the wrong q-Stokes matrix. The paper does contain the correct-looking formula in the proof, suggesting a typographical error in the statement, but as submitted the central claim is not correct. This warrants requiring a correction of the theorem statements before acceptance, so I recommend conditional acceptance rather than outright rejection, because the underlying method and the proof's internal formula (50) are recoverable and verifiable.","tokens_in":38105,"tokens_out":29210,"duration_ms":265663,"concrete_test":"Fix n=3 and take A = [[0,1,0],[2,0,0],[0,1,3]] with a generic q (e.g., q=0.5) and generic λ, μ, z. Compute (b_q)_1 in three ways: (i) using the printed Theorem 1.1 denominator-minor formula; (ii) using the proof's equation (50), i.e., -Σ_j b_j^{(2)}(P^{-1})_{j1} times the inner sum, with b_j^{(2)} and P^{-1} from Proposition 3.4; (iii) directly from S_q = U_q(z,λ) U_q(z,μ)^{-1} using the explicit U_q entries in Theorems 3.8 and 3.9. Methods (ii) and (iii) should agree; if method (i) disagrees, the theorem statement's explicit formula is wrong and needs correction.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central result is the explicit q-Stokes matrix in Theorem 1.1 (restated as Theorem 3.12). In the displayed formula, the k-th entry of b_q contains the factor Δ_{1..n-2,n}^{1..n-1}(A-λ_j I) / [Δ_{1..\\hat{k}..n-1}^{1..n-2}(A-λ_j I) · ∏_{l≠j}(λ_l-λ_j) · ∏_{l=1}^{n-2}(λ_l-λ_j)]. But the proof of Theorem 3.12 first derives (b_q)_k = -Σ_j b_j^{(n-1)} (P^{-1})_{jk} [inner sum] (equation (50)). By Proposition 3.4, b_j^{(n-1)} is proportional to Δ_{1..n-2,n}^{1..n-1}(A-λ_j I), and (P^{-1})_{jk} is proportional to Δ_{1..\\hat{k}..n-1}^{1..n-2}(A-λ_j I). Thus the two minors multiply in the numerator. Unless the second minor equals its reciprocal (i.e., equals ±1), the printed denominator form is not algebraically equivalent to (50). For a generic 3×3 example, the two expressions differ, so Theorem 1.1 as stated is incorrect even though equation (50) appears to contain the correct formula. This directly undermines the paper's main claim of explicitly evaluating the q-Stokes matrix.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the confluent hypergeometric q-difference system D_q F_q(z) = (E_nn + A/z) F_q(z) and claims an explicit formula for its q-Stokes matrix. The main technical tool is a connection formula, proved via q-Borel resummation and a confluence limit from Thomae-type formulas, for the basic hypergeometric series nphi_{n-1}(a_1,...,a_{n-1},0;b_1,...,b_{n-1};q,z). The paper also shows that, as q tends to 1, the q-Stokes matrix and the connection matrix recover the known Stokes matrices and connection matrices of the corresponding differential system. The central result is Theorem 1.1 (restated as Theorem 3.12), giving the lower block b_q in the q-Stokes matrix.","tokens_in":38382,"tokens_out":4784,"duration_ms":48917,"significance":"If correct, the result would be a substantial contribution: it provides complete explicit Stokes data for a family of irregular confluent hypergeometric q-difference equations, going beyond previously known low-dimensional cases. The connection formula for nphi_{n-1} with one zero numerator parameter is also of independent interest, and the q-to-1 limit connects the new formulas to the classical differential Stokes phenomenon. The paper is transparent about the ingredients: it uses q-Borel summability results from the literature and derives the Stokes matrix by multiplying connection matrices rather than by imposing the answer. However, the main theorem as printed contains an algebraic error in the placement of a minor, and one identification step in the proof of the connection formula rests on an unstated uniqueness argument. These issues must be addressed before the central claim can be accepted.","major_comments":[{"comment":"The displayed formula for (b_q)_k is not algebraically equivalent to the formula derived in the proof. In the proof, equation (50) gives (b_q)_k = -\\sum_j b_j^{(n-1)} (P^{-1})_{jk} [inner sum]. By Proposition 3.4, b_j^{(n-1)} is proportional to \\Delta_{1,...,n-2,n}^{1,...,n-1}(A-\\lambda_j^{(n-1)} I) and (P^{-1})_{jk} is proportional to \\Delta_{1,...,\\hat{k},...,n-1}^{1,...,n-2}(A-\\lambda_j^{(n-1)} I). Their product therefore contains the second minor in the numerator. The theorem as printed places \\Delta_{1,...,\\hat{k},...,n-1}^{1,...,n-2} in the denominator. These two expressions differ for a generic 3x3 matrix, so the statement of Theorem 1.1 is incorrect as written. Since this is the paper's main claim, the display must be corrected (the proof's equation (50) appears to contain the correct formula).","section":"Theorem 1.1 / §3.6, Theorem 3.12"},{"comment":"The identification \\psi(z) = h_q(z)_n is asserted by 'Comparing with (11) and (12)' without proving the needed uniqueness. The function \\psi(z) is shown to be a power series in 1/z with constant term 1, and \\psi(z) z^{\\sum \\log_q(a_l/b_l)}/(z;q)_\\infty is shown to satisfy equation (5). To conclude \\psi = h_q(z)_n, one must prove that the equation (5) has at most one solution of this precise asymptotic form with constant term 1. This uniqueness is plausible and follows from the recursive determination of the coefficients in (12), but the paper does not isolate or prove it. Since Theorem 1.2 and hence the q-Stokes computation depend on this identification, the gap should be filled explicitly.","section":"§2.4, equation (28)"}],"minor_comments":[{"comment":"The word 'hypergeomtric' should be 'hypergeometric'.","section":"Abstract"},{"comment":"In the sentence 'from (6) and Lemma 2.22, we get (here for convenience, we denote b_n = q)', the notation b_n = q is introduced but the subsequent displayed formula would be clearer if the special role of b_n were stated before the limit is taken.","section":"§2.4, proof of Theorem 2.26"},{"comment":"The phrase 'note that we denote b_n = q here' is easy to miss; since b_n plays a different role from the parameters b_1,...,b_{n-1}, it would help to define this convention in a displayed line.","section":"§2.4, Corollary 2.27"},{"comment":"The sentence 'To prove Proposition 4.6, we need the following lemma' should refer to Proposition 4.16, not Proposition 4.6.","section":"§4.2, proof of Proposition 4.16"}],"recommendation":"major_revision","confidential_remarks":"The error in the displayed formula of Theorem 1.1 appears to be a typographical slip, since the proof's equation (50) together with Proposition 3.4 yields the corrected numerator placement. The uniqueness gap in the proof of Theorem 2.26 is more substantive and needs an explicit argument. Both are within the manuscript's scope to fix. I do not see grounds for rejection, but the main theorem must be corrected and the uniqueness point proved before the paper is publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know up front. First, the paper delivers something genuinely new: a general-n connection formula for nφ_{n−1}(a_1,...,a_{n−1},0;b_1,...,b_{n−1};q,z), derived via q-Borel resummation and a confluence argument, and then explicit q-Stokes matrices for the system D_q F_q = (E_nn + A/z)F_q. The n=2 case reproduces Zhang's earlier result, which is a good sanity check. Second, the main theorem as printed is wrong. The stress-test note checks out: in Theorem 1.1/3.12, the minor Δ_{1..\\hat{k}..n−1}^{1..n−2}(A−λ_j I) sits in the denominator of (b_q)_k, but the proof's own equation (50), obtained by conjugating with diag(P_{n−1},1), has that minor in the numerator, multiplied by the other minor Δ_{1..n−2,n}^{1..n−1}. I verified the signs and denominators; the printed expression is not algebraically equivalent to (50). So the central explicit formula is misstated. The good news is that the proof contains the correct formula, and the mistake looks like a copy-paste error rather than a deep flaw.\n\nWhat the paper does well: the connection formula is a real extension of prior work, the q-Borel machinery is standard and cited correctly, and the q→1 limit is a useful check, though the abstract overclaims a bit: the recovery of Stokes matrices is proved for S_− under sector conditions (80)-(81), not for both Stokes matrices unconditionally. There is also a small gap in the proof of Theorem 2.26: the identification ψ(z)=h_q(z)_n is asserted by 'comparing with (11) and (12)' without an explicit uniqueness argument. That is a minor expository gap; it should be easy to fill. The block form assertion in Theorem 3.10 is also a bit compressed.\n\nOverall, the mathematical strategy is sound and the paper deserves a serious referee. The referee should require the correction of Theorem 1.1, a sharper statement of the q→1 result, and a couple of added details. After those fixes, it would be a solid contribution to the analytic q-difference literature.","headline":"Genuinely new connection formula, but Theorem 1.1's q-Stokes formula has a misplaced minor that must be corrected before the paper can be used as stated.","tokens_in":38982,"tokens_out":4751,"would_cite":false,"duration_ms":45652,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["39A13","33D15","34M40"],"pacs":[],"model":"deepseek-v4-flash","headline":"The authors prove a connection formula for confluent basic hypergeometric series and use it to compute the q-Stokes matrices of an irregular q-difference system explicitly.","keywords":["q-Stokes matrices","q-Borel resummation","basic hypergeometric series","confluent hypergeometric q-difference equations","connection formula","q-theta functions","Stokes phenomenon","q to 1 limit"],"falsifier":"Take $n=2$ and a concrete $2\\times2$ matrix $A$ satisfying the hypotheses, choose $q=1/2$, and compute the single nontrivial entry $(b_q)_1$ from Theorem 1.1. Independently evaluate both sides of $F_q^{(\\infty)}(z,\\mu;E_{22},A)=F_q^{(\\infty)}(z,\\lambda;E_{22},A)S_q(z,\\lambda,\\mu;E_{22},A)$ at a point off the relevant $q$-spirals by truncating the $q$-Borel sums; any disagreement disproves the formula. A more structural check is to prove or disprove the uniqueness used in (28): if the equation (5) admits a second formal solution $\\sum c_k z^{-k}$ with $c_0=1$ distinct from $h_q(z)_n$, then the connection formula's coefficient identification fails.","tokens_in":37861,"feed_emoji":"🌀","tokens_out":16092,"duration_ms":137656,"temperature":0.7,"pith_summary":"The paper proves a connection formula for the confluent basic hypergeometric series ${}_n\\phi_{n-1}(a_1,\\ldots,a_{n-1},0;b_1,\\ldots,b_{n-1};q,z)$, a solution of the confluent hypergeometric $q$-difference equation, by $q$-Borel resummation. The main application is an explicit evaluation of the $q$-Stokes matrix, the transition matrix between two meromorphic solutions that share one formal asymptotic expansion but are summed along different $q$-spiral directions, for the irregular system $D_qF_q(z)=(E_{nn}+A/z)F_q(z)$. Under nonresonance conditions on the eigenvalues of $A$, the matrix is lower block triangular and its last row is given by a closed formula in $q$-Pochhammer symbols, $q$-$\\theta$ functions, and minors of $A$. This supplies complete explicit Stokes data for this family of irregular confluent hypergeometric $q$-difference systems. As $q\\to1$, the formula recovers the known Stokes matrices of the corresponding confluent hypergeometric differential system, so the result is a genuine $q$-deformation of the classical Stokes phenomenon.","feed_headline":"Explicit q-Stokes matrices computed for irregular q-difference systems","feed_subtitle":"Full formula for the Stokes row, with q→1 limit matching classical differential Stokes matrices.","key_machinery":"The carrier of the argument is $q$-Borel resummation: the $[\\,\\lambda;q\\,]$-sum of a divergent formal power series $\\sum a_n z^{-n}$ is obtained by applying the $q$-Borel transform $\\widehat B_{q;1}$ and then the $q$-Laplace summation $L_{q;1}^{[\\lambda;q]}$, producing a meromorphic solution whose poles lie on the $q$-spiral $[-\\lambda^{-1};q]$. The notation ${}_n f_{n-2}(a;b;\\lambda;q,z)$ denotes the resulting sum of a divergent ${}_n\\phi_{n-2}$, and the key confluence lemmas (Proposition 2.25 and Theorem 2.26) pass from ${}_n\\phi_{n-1}$ with a parameter $-\\lambda q^m$ to the confluent series with a zero upper parameter by taking $m\\to\\infty$ along the $q$-spiral. The delicate step is the identification $\\psi(z)=h_q(z)_n$ in (28): the function $\\psi(z)$ produced by the confluence is shown to satisfy the same $q$-difference equation and to have a power-series expansion in $1/z$ with constant term $1$, which matches the recursion (12) defining $h_q(z)_n$; the comparison is what transfers the explicit coefficients to the final formula.","core_discovery":"On the paper's own terms, the central discovery is Theorem 1.1. For $A\\in\\mathfrak{gl}_n$ with $\\mu_i-\\mu_j\\notin\\mathbb{Z}$, $\\nu_i-\\nu_j\\notin\\mathbb{Z}$, and $\\lambda^{(n-1)}_j\\neq\\lambda^{(n-2)}_\\ell$, the $q$-Stokes matrix of $D_qF_q(z)=(E_{nn}+A/z)F_q(z)$ is $\\begin{pmatrix}\\mathrm{Id}_{n-1}&0\\\\ b_q&1\\end{pmatrix}$, where the row vector $b_q$ is a double sum over products of $q$-Pochhammer symbols $(a;q)_\\infty$, ratios of $q$-$\\theta$ functions $\\theta_q$ evaluated at $\\lambda z$ and $\\mu z$, powers $z^{\\nu_j}/z^{\\nu_n+1}$, and minors $\\Delta^{1,\\ldots,n-2,n}_{1,\\ldots,n-1}(A-\\lambda^{(n-1)}_j\\mathrm{Id}_n)$ of $A$. The route is a confluence limit that converts the classical non-confluent connection formula for ${}_n\\phi_{n-1}$ into a connection formula (Theorem 1.2) for the confluent series with an upper parameter equal to $0$; this connection formula is then fed into the diagonalized system (33), yielding the connection matrix and then the $q$-Stokes matrix. The paper further proves in Proposition 4.16 that, after the factor $e^{-2\\pi i\\delta_q^{(n-1)}(A_{n-1})}$, the $q$-Stokes matrix tends as $q\\to1$ to the classical Stokes matrix $S_-(E_{nn},A_{n-1})$ of the differential system $F'=(E_{nn}+A/z)F$.","pith_inferences":["The same confluence route should give explicit connection matrices for other degenerate basic hypergeometric series, such as ${}_n\\phi_n$ with a zero parameter or ${}_n\\phi_m$ with a denominator parameter tending to zero; the paper's Remarks 2.28 and 3.13 point at some of these but leave them to later work.","The explicit $b_q$ formula is numerically checkable for small $n$: every ingredient is a computable $q$-series or $q$-theta function, so one can compare the two sides of the $q$-Stokes definition at a point away from the $q$-spirals.","If the missing uniqueness step in (28) were to fail, the formulas would still be natural candidates but would need a different proof; a formal uniqueness theorem for solutions of (5) with prescribed asymptotic constant term would close the gap.","Reading the $q\\to1$ limit backwards suggests that the $q$-Stokes matrix interpolates the classical Stokes matrix along the $q$-deformation, which could be used to transport isomonodromic data between the $q$-world and the differential world."],"forward_implications":["The connection matrix $U_q(z,\\lambda;E_{nn},A)$ for the same system is explicit (Theorem 3.8), and its inverse is explicit as well (Theorem 3.9), so the full transition data between solutions around $0$ and around $\\infty$ is available.","Taking $q\\to1$ in Theorem 1.1 via Proposition 4.16 reproduces the classical Stokes matrices of $F'=(E_{nn}+A/z)F$ once the diagonal factor $e^{-2\\pi i\\delta_q^{(n-1)}(A_{n-1})}$ is removed; the $q$-Stokes matrix is therefore a genuine $q$-analogue of the differential Stokes matrix.","For $n=2$, Theorem 1.2 reduces to the known confluent connection formula for ${}_2\\phi_1(a_1,0;b_1;q,z)$, recovering a previously established case and validating the confluence procedure.","The explicit formulas give the $q$-Stokes data needed for Riemann-Hilbert and isomonodromy problems attached to this irregular confluent family, in the direction the paper's introduction points."],"supporting_citations":[{"why":"Supplies the q-Borel summation theory for divergent basic hypergeometric series and the fundamental solutions used in Proposition 2.20.","marker":"[1]"},{"why":"Supplies the formal and Borel-summed solutions of the limiting differential system (68), the objects that the q to 1 limits in Section 4 are compared with.","marker":"[4]"},{"why":"Gives the fundamental system of meromorphic solutions of the basic hypergeometric q-difference equation (9) around infinity, used in Proposition 2.4.","marker":"[5]"},{"why":"Supplies the confluence of meromorphic solutions of q-difference equations, the framework for the limiting process (17)-(18) and the q-Borel summability of the system's formal solutions.","marker":"[6]"},{"why":"Provides the standard q-Pochhammer and q-theta identities, including the non-confluent connection formula framework used in Lemma 2.22.","marker":"[8]"},{"why":"Gives the explicit gauge transformation of Section 3.2 and the companion Stokes-matrix result that the present paper extends to the q case.","marker":"[15]"},{"why":"Foundational for q-Borel resummation and q-asymptotics, justifying the existence of the meromorphic solutions whose transitions define the q-Stokes matrices.","marker":"[30]"},{"why":"The classical connection formula for non-confluent n phi n-1 is the starting point subjected to confluence in Theorem 2.26.","marker":"[37]"},{"why":"Supplies Lemma 2.24 for q-Pochhammer limits and the n=2 confluent connection formula that Theorem 1.2 recovers.","marker":"[45]"}],"fun_headline_variants":["Explicit q-Stokes matrices for confluent q-difference equations","q-Stokes matrices: full formula and classical limit","Explicit q-Stokes matrices via confluence, with q→1 limit","Closed form for q-Stokes matrices, matching classical differential limit","q-Stokes matrices: explicit formula and q→1 recovery"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof hinges on the unproved uniqueness of a solution of the confluent $q$-difference equation with a prescribed asymptotic expansion in powers of $1/z$ and constant term $1$: the confluence limit $\\psi(z)$ is identified with the recursively defined series $h_q(z)_n$ purely by comparing their asymptotic forms at the step marked 'Comparing with (11) and (12)', and if that identification is not forced, the connection formula and the $q$-Stokes matrices would not be established.","fun_headline_variants_meta":{"raw":{"variants":["Explicit q-Stokes matrices for confluent q-difference equations","q-Stokes matrices: full formula and classical limit","Explicit q-Stokes matrices via confluence, with q→1 limit","Closed form for q-Stokes matrices, matching classical differential limit","q-Stokes matrices: explicit formula and q→1 recovery"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000744,"raw_usage":{"total_tokens":3369,"prompt_tokens":1046,"completion_tokens":2323,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":662,"completion_tokens_details":{"reasoning_tokens":2235}},"tokens_in":662,"tokens_out":2323,"duration_ms":19230,"temperature":1.0,"reasoning_tokens":2235,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T23:38:05.280633+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take $n=2$ and a concrete $2\\times2$ matrix $A$ satisfying the hypotheses, choose $q=1/2$, and compute the single nontrivial entry $(b_q)_1$ from Theorem 1.1. Independently evaluate both sides of $F_q^{(\\infty)}(z,\\mu;E_{22},A)=F_q^{(\\infty)}(z,\\lambda;E_{22},A)S_q(z,\\lambda,\\mu;E_{22},A)$ at a point off the relevant $q$-spirals by truncating the $q$-Borel sums; any disagreement disproves the formula. A more structural check is to prove or disprove the uniqueness used in (28): if the equation (5) admits a second formal solution $\\sum c_k z^{-k}$ with $c_0=1$ distinct from $h_q(z)_n$, then the connection formula's coefficient identification fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the q-Borel summation theory for divergent basic hypergeometric series and the fundamental solutions used in Proposition 2.20."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the formal and Borel-summed solutions of the limiting differential system (68), the objects that the q to 1 limits in Section 4 are compared with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the fundamental system of meromorphic solutions of the basic hypergeometric q-difference equation (9) around infinity, used in Proposition 2.4."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the confluence of meromorphic solutions of q-difference equations, the framework for the limiting process (17)-(18) and the q-Borel summability of the system's formal solutions."},{"cited_title":"Gasper and M","cited_arxiv_id":null,"evidence_quote":"Provides the standard q-Pochhammer and q-theta identities, including the non-confluent connection formula framework used in Lemma 2.22."},{"cited_title":"Explicit evaluation of the Stokes matrices for certain quantum confluent hypergeometric equations","cited_arxiv_id":"2401.13911","evidence_quote":"Gives the explicit gauge transformation of Section 3.2 and the companion Stokes-matrix result that the present paper extends to the q case."},{"cited_title":"Ramis, J","cited_arxiv_id":null,"evidence_quote":"Foundational for q-Borel resummation and q-asymptotics, justifying the existence of the meromorphic solutions whose transitions define the q-Stokes matrices."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The classical connection formula for non-confluent n phi n-1 is the starting point subjected to confluence in Theorem 2.26."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies Lemma 2.24 for q-Pochhammer limits and the n=2 confluent connection formula that Theorem 1.2 recovers."}],"review_version":1}