{"id":"6aa5247f-d3a6-4e8c-8f14-19bd04a58d2e","arxiv_id":"2507.07423","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Higher Hida theory, including an interpolated Serre duality pairing, is constructed for Drinfeld modular curves.","lead":"This paper constructs a p-adic interpolation of higher-degree cohomology of Drinfeld modular forms, a function field analog of modular forms. It is a technical step toward p-adic L-functions and modularity results in the function field setting.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The duality interpolation in §7.3 drops the (−2D) twist: the displayed identification D(ωκun) ≅ ωκun ⊗_{Λ,d} Λ is false as written, and the module N and the pairing need a corrected twist.","rationale":"Read in good faith, the paper is a plausible adaptation of Boxer–Pilloni, and the construction of the family ωκun via the Igusa tower is a reasonable route. The strongest claim needs four ingredients: a universal family, projectors e(U_p) and e(F), a dual family whose classical fibers realize Serre duality, and finite generation. The weakest link is the third: the identification in §7.3 of D(ωκun) with ωκun ⊗_{Λ,d} Λ silently loses the divisor (−2D) that appears two lines earlier and in the stated theorem. This is not cosmetic: if D(ωκun) is taken literally, its f_k-specialization is ω^{2−k}, so the claimed N-fiber and the interpolation of Serre duality do not follow. The likely repair is to write (ωκun ⊗_{Λ,d} Λ)(−2D), but the proof also depends on the asserted isomorphism ω ≅ ω^D on X^ord, which the authors themselves flag as delicate (Remark 4.5) and which is not proved. The finite-generation issue is real, but it is secondary to the written proof of the stated theorem. I therefore do not move the reader's verdict: the paper should remain CONDITIONAL until the twist is corrected and the ω ≅ ω^D identification on X^ord is justified.","tokens_in":22540,"tokens_out":16254,"duration_ms":175131,"concrete_test":"Recompute the specialization of both sides of the §7.3 isomorphism at f_k. Using Ω^1_{X/A_p} ≅ ω ⊗ ω^D(−2D) and Hom(ωκun, ...) ≅ (ωκun)^{-1}, the specialization of D(ωκun) is ω^{1−k} ⊗ ω^D(−2D), while the specialization of the written ωκun ⊗_{Λ,d} Λ is ω^{2−k}. Check whether inserting (−2D) into the §7.3 isomorphism and into Theorem 7.7's module makes the specializations match the theorem and Corollary 5.4(2). This is a local calculation on X^ord and settles whether the omission is a harmless typo or signals a wrong dualizing family.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires the isomorphism in §7.3 identifying the dualizing family with ωκun twisted by d. Step by step, D(ωκun) = Ω^1_{X/A_p} ⊗ Hom(ωκun, Λ ⊗ O_Xord) ≅ ω ⊗ ω^D(−2D) ⊗ Hom(ωκun, ...) by the Kodaira–Spencer relation, then ≅ ω^2(−2D) ⊗ Hom(ωκun, ...) if ω ≅ ω^D on X^ord. But the displayed simplification then says D(ωκun) ≅ ωκun ⊗_{Λ,d} Λ, silently omitting the divisor (−2D). Specializing at f_k, the written object gives ω^{2−k}, whereas the theorem and Corollary 5.4(2) require ω^{1−k} ⊗ ω^D(−2D). Since N is built from e(F)H^1_c of this object and the perfect pairing is claimed to interpolate Serre duality, this omission is load-bearing. The same spot is where the paper's own warning about lack of auto-duality (Remark 4.5) is closest to the argument: the identification ω ≅ ω^D on X^ord is asserted without proof in Corollary 5.4(2) and used again here. If ω ≅ ω^D holds, the corrected identity should be (ωκun ⊗_{Λ,d} Λ)(−2D); if it fails, N and the pairing need a different construction.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a higher Hida theory for Drinfeld modular curves, following the Boxer–Pilloni template. The authors construct a universal family ω^{κ_un} of line bundles over the ordinary locus using an Igusa tower and the Hodge–Tate–Taguchi map, define Hecke correspondences T_p, U_p, F, prove local finiteness and control theorems for their projectors, and define Λ-modules M and N whose classical specializations recover e(T_p)H^0(X,ω^k) and e(T_p)H^1(X,ω^{1−k}⊗ω^D(−2D)). They then construct a Serre-duality pairing M×N→Λ. The main theorem is summarized in the Introduction and is meant to be proved by Theorems 7.5 and 7.7.","tokens_in":22758,"tokens_out":8946,"duration_ms":100531,"significance":"If the announced results are correct, this is a meaningful contribution: it extends higher Hida theory to the function-field setting, handles a non-Noetherian Iwasawa algebra Λ, and accommodates the lack of canonical self-duality of Drinfeld modules by systematically using the dual bundle ω^D. The paper also contains a self-contained proof of Serre–Tate coordinates for Drinfeld modules and proposes a concrete interpolation of Serre duality. The constructions are canonical and contain no fitted parameters, and the paper correctly identifies the places where the Drinfeld situation differs from the elliptic modular case.","major_comments":[{"comment":"The displayed isomorphism D(ω^{κ_un}) ≅ ω^{κ_un}⊗_{Λ,d}Λ is missing the (−2D) twist. From the preceding line one has D(ω^{κ_un}) ≅ ω^2(−2D)⊗Hom(ω^{κ_un},Λ⊗O_{X^ord}), which with d(t)=t^2(κ_un(t))^{-1} specializes at weight k to ω^{2−k}(−2D), i.e. to (ω^{κ_un}⊗_{Λ,d}Λ)(−2D). This is exactly the combination needed for Theorem 7.7 and for the module N defined in the Introduction, but the displayed simplification omits the divisor. As written, the isomorphism contradicts the subsequent specialization and the statement of the theorem; it must be corrected to include (−2D).","section":"§7.3"},{"comment":"The isomorphism ω ≅ ω^D on the ordinary locus is asserted without proof in the proof of Corollary 5.4(2) ('and the isomorphism ω ∼= ωD in Xord') and again in §7.3 before the simplification D(ω^{κ_un}) ≅ ω^{κ_un}⊗_{Λ,d}Λ. This is load-bearing: it is used to identify the Serre-dual family and to justify the target H^1(X,ω^{1−k}⊗ω^D(−2D)). In view of Remark 4.5, which explicitly warns that Drinfeld modules are not canonically self-dual, this is not a standard fact and needs a proof or a precise reference. If the isomorphism fails, the module N and the pairing require a different twist.","section":"Corollary 5.4(2) and §7.3"},{"comment":"The local finiteness of F (and similarly of U_p) is not actually proved. The text says 'The next point is to check that F is locally finite... The case n=1 follows... For the induction step we can use the exact sequence ... Now, using exactness properties of the notion of locally finite, one can deduce the properties for n+1 from n and H^1_c(...)'. The existence of the projectors e(F) and e(U_p) depends on this local finiteness, so the induction and the quoted 'exactness properties' must be supplied, or a precise reference to [4] must be given with verification that the non-Noetherian adaptation still works.","section":"Theorem 7.5"},{"comment":"The finite-type property of the Λ-modules M and N is part of the announced main theorem, but it is not proved in the body. The sentence 'Further analysis in the arguments also leads to the finite type property' is not a proof, and since Λ is non-Noetherian, finite generation is not automatic. A proof of finite type (or a precise reference) is also needed for the perfection argument of the pairing in Theorem 7.7, where the density reduction to classical weights is not by itself sufficient to conclude that the global Λ-pairing is perfect.","section":"Introduction and Theorems 7.5/7.7"}],"minor_comments":[{"comment":"There is a typo: 'whcih' should be 'which'.","section":"§2.1"},{"comment":"The notation D(ω^{κ_un}) is used inconsistently: in §4.2, D is the dualizing functor RHom(−,q^!A_p), which is concentrated in degree 1, whereas in §7.3 D(ω^{κ_un}) denotes Ω^1⊗Hom(ω^{κ_un},Λ⊗O) without a shift. Please clarify the convention.","section":"§7.3"},{"comment":"The bibliography entry [dS16] appears out of alphabetical order and does not seem to be cited in the text; either cite it or remove it.","section":"References"},{"comment":"The operator ⟨ϖ⟩ is used before it is defined; please state explicitly its action on the prime-to-p level structure earlier in the paper.","section":"§6.3"},{"comment":"The notation 'Ec' in 'Ec[p∞]' is not introduced; presumably it means the connected part of E[p∞]. Please define it.","section":"§3"}],"recommendation":"major_revision","confidential_remarks":"The twist omission in §7.3 looks like a correctable typo, but the unproved isomorphism ω ≅ ω^D on the ordinary locus is potentially a deeper issue, since the paper itself emphasizes the absence of canonical auto-duality. The missing proofs of local finiteness and finite type are substantial gaps relative to the announced theorem, but they appear fillable within the manuscript's scope rather than fatal. I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. The main theorem is a genuine extension: first higher Hida theory in the function field setting, interpolating degree-one cohomology and Serre duality for Drinfeld modular curves. The construction follows Boxer–Pilloni, which is a strength—the paper adapts a known template rather than inventing fragile machinery. The Serre–Tate coordinates for Drinfeld modules (Theorem 3.3) are also a useful contribution, since the proof was folklore. The local analysis of T_p on the supersingular locus is careful, and the condensed-mathematics setup for H^1_c is appropriate.\n\nThe soft spots are concentrated in the duality part. In §7.3, the written isomorphism D(ωκun) ≅ ωκun ⊗_{Λ,d} Λ drops the (−2D) divisor that appeared one line earlier. As written that gives ω^{2−k} at weight k, whereas the theorem and the proof's own specialization step require ω^{1−k}⊗ω_D(−2D). The introduction and Theorem 7.7 use the twisted version, so this is probably a slip rather than a fatal flaw, but it needs fixing before the pairing argument is complete. Second, the isomorphism ω ≅ ω^D on X^ord is asserted without proof in Corollary 5.4(2) and again in §7.3. Given Remark 4.5, this is exactly the kind of fact that needs a reference or a short argument. Third, Theorem 7.5's local finiteness induction is sketched, not proved, and the finite type of M and N over the non-Noetherian Λ is stated without a proof in the body.\n\nThe citation pattern looks fine: [15] is the published degree-zero theory, used correctly as input, not as a hidden circular step. The paper is honest about what it adapts and what it leaves to the reader.\n\nThis deserves a serious referee. The skeleton is sound and the project is important to function-field arithmetic and to higher Hida theory in general. The referee should push on the (−2D) twist and the ω≅ω^D claim. I would not cite the full theorem in my own work until a corrected version appears, but I'd bring it to a reading group and I'd vote to send it out.","headline":"Genuine first higher Hida theory for Drinfeld modular curves on a solid template, but the duality section has a missing divisor twist and two load-bearing proofs deferred.","tokens_in":23392,"tokens_out":9272,"would_cite":false,"duration_ms":94834,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11F52","11F33","11G09"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper establishes higher Hida theory for Drinfeld modular curves: two finite modules over the Iwasawa algebra interpolate ordinary degree-zero and degree-one cohomology, with a perfect pairing over the family.","keywords":["higher Hida theory","Drinfeld modular curves","Drinfeld modular forms","Serre duality interpolation","Iwasawa algebra","ordinary locus","Hodge–Tate–Taguchi map","Hecke operators"],"falsifier":"Look for a rank-two ordinary Drinfeld module over $A_p$ for which the Hodge–Tate–Taguchi map $HTT: (H^{\\mathrm{can}})^D \\to \\omega$ is not an isomorphism or has a non-trivial zero; the existence of even one such fibre would make $\\omega\\cong\\omega^D$ fail on $X^{\\mathrm{ord}}$, invalidating Corollary 5.4(2) and the identification $D(\\omega^{\\kappa^{\\mathrm{un}}})\\cong\\omega^{\\kappa^{\\mathrm{un}}}\\otimes_{\\Lambda,d}\\Lambda$ that produces the perfect pairing.","tokens_in":22243,"feed_emoji":"🧮","tokens_out":14742,"duration_ms":134442,"temperature":0.7,"pith_summary":"This paper transplants higher Hida theory from the number-field setting to Drinfeld modular curves, deforming not only ordinary Drinfeld modular forms (degree-zero cohomology) but also their degree-one coherent cohomology. The main theorem produces two finite-type modules $M$ and $N$ over the Iwasawa algebra $\\Lambda = A_p[[A_p^\\times]]$ whose specializations at every integer weight $k \\geq 3$ recover the ordinary parts $e(T_p)H^0(X,\\omega^k)$ and $e(T_p)H^1(X,\\omega^{1-k}\\otimes \\omega^D(-2D))$, together with a perfect pairing $M\\times N\\to\\Lambda$ interpolating Serre duality. This matters because higher Hida theory has powered arithmetic applications over number fields, and the function-field analogue has so far existed only in degree zero. The proof proceeds by interpolating the bundles $\\omega^k$ through the Hodge–Tate–Taguchi map on the ordinary locus and then using projectors $e(U_p)$ and $e(F)$ to isolate the ordinary parts.","feed_headline":"Drinfeld modular curves gain higher Hida theory","feed_subtitle":"Two Iwasawa modules interpolate weight-k forms and their Serre duals at every k≥3.","key_machinery":"The load-bearing objects are the Igusa tower $Ig = \\mathrm{Isom}_{X^{\\mathrm{ord}}}(A_p, T_p((H^{\\mathrm{can}})^D))$ and the universal weight $\\kappa^{\\mathrm{un}}: A_p^\\times\\to\\Lambda^\\times$, which together define the interpolating line bundle $\\omega^{\\kappa^{\\mathrm{un}}} = (O_{Ig}\\widehat{\\otimes}\\,\\Lambda)^{A_p^\\times}$. On this family the paper constructs two locally finite operators $U_p$ and $F$, with associated projectors $e(U_p)$ and $e(F)$, that specialize at integer weights to the classical Hecke operator $T_p$. The second ingredient is the Kodaira–Spencer isomorphism $\\omega\\otimes\\omega^D \\cong \\Omega^1_{X/A_p}(2D)$; combined with the identification $\\omega\\cong\\omega^D$ on $X^{\\mathrm{ord}}$, it identifies the dualizing family $D(\\omega^{\\kappa^{\\mathrm{un}}})$ and yields the perfect pairing.","core_discovery":"The central claim, stated as Theorems 7.5 and 7.7, is that higher Hida theory exists for the Drinfeld modular curve $X$: there are finite-type $\\Lambda$-modules $M$ and $N$ with Hecke action such that for every integer $k\\geq 3$, $$M\\otimes_{\\Lambda,k}A_p \\cong e(T_p)$H^{0}$(X,\\omega^k),\\qquad N\\otimes_{\\Lambda,k}A_p \\cong e(T_p)$H^{1}$(X,\\$omega^{{1-k}}$\\otimes\\omega^D(-2D)),$$ and $M\\times N\\to\\Lambda$ is a perfect pairing interpolating Serre duality. The degree-zero statement extends the earlier Hida theory for Drinfeld modular forms; the degree-one statement is the genuinely new input, obtained by defining compact-support cohomology $H^1_c(X^{\\mathrm{ord}},\\omega^{\\kappa^{\\mathrm{un}}})$ through condensed mathematics and cutting out the ordinary part with the projectors $e(U_p)$ and $e(F)$. A function-field version of the Serre–Tate coordinates theorem is proved along the way and is used to show that the Hecke correspondence $T_p$ is integrable on the ordinary locus.","pith_inferences":["Beyond the paper, replacing the fragile identification $\\omega\\cong\\omega^D$ by an explicit twist with the dual bundle should produce a parallel construction that does not rely on Drinfeld modules being self-dual.","Beyond the paper, the same locally finite projectors and condensed six-functor formalism should yield a higher Coleman theory (overconvergent families) for Drinfeld modular curves, since the projectors are built from locally finite operators whose limit behaviour is already controlled.","Beyond the paper, the profinite filtration used because $\\Lambda$ is non-Noetherian may adapt to other function-field Shimura varieties, where Noetherianity of the Iwasawa algebra also fails."],"forward_implications":["For every $k\\geq 3$, the ordinary part of the space of Drinfeld modular forms of weight $k$ is the specialization of one fixed $\\Lambda$-module $M$, so the weights are controlled uniformly by a single family.","The degree-one module $N$ specializes to $e(T_p)H^1(X,\\omega^{1-k}\\otimes\\omega^D(-2D))$, giving higher-degree coherent cohomology the same deformation-theoretic control that Hida theory gives to modular forms.","The perfect pairing $M\\times N\\to\\Lambda$ specializes to the classical Serre duality pairing at each weight; in particular, the Serre dual of $T_p$ is, up to diamond operators, $U_p$, matching the elliptic modular curve situation.","The proof includes a Serre–Tate coordinates theorem for ordinary Drinfeld modules, giving local coordinates on the deformation ring and a description of the correspondence $T_p$ near the ordinary locus that is used to prove its integrability."],"supporting_citations":[{"why":"Supplies the blueprint for higher Hida theory: locally finite operators, projectors, and the pairing argument on the modular curve.","marker":"[4]"},{"why":"Supplies the canonical subgroups, the Hodge–Tate–Taguchi map, and the dual Drinfeld module bundle $\\omega^D$.","marker":"[11]"},{"why":"Supplies Cartier–Taguchi duality and the pairing with the Carlitz module used throughout the paper.","marker":"[23]"},{"why":"Supplies the degree-zero Hida theory for Drinfeld modular forms that this paper extends, including the congruence $U_p\\equiv T_p$ modulo $\\varpi$.","marker":"[15]"},{"why":"Supplies the level structure, compactification, and canonical subgroup results for the Drinfeld modular curve.","marker":"[12]"},{"why":"Supplies the Kodaira–Spencer isomorphism used to compute Serre duality and the dual of $T_p$.","marker":"[13]"},{"why":"Supplies the six-operations formalism and the identification of compact-support cohomology used to define $H^1_c$.","marker":"[5]"},{"why":"Supplies the condensed-mathematics six-functor formalism in which $H^1_c(X^{\\mathrm{ord}},\\omega^{\\kappa^{\\mathrm{un}}})$ is defined.","marker":"[7]"},{"why":"Supplies the Serre–Tate local moduli argument used in Section 3 to prove the function-field Serre–Tate coordinates theorem.","marker":"[24]"}],"fun_headline_variants":["Higher Hida theory for Drinfeld modular curves, with Serre duality","New degree-one Hida theory on Drinfeld modular curves","Drinfeld modular curves: higher Hida theory interpolates Serre duality","Higher Hida theory for Drinfeld curves now includes Serre duality","Condensed mathematics enables higher Hida theory on Drinfeld curves"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction assumes that on the ordinary locus the modular-forms line bundle $\\omega$ is isomorphic to the dual bundle $\\omega^D$; this identification is used without proof in Corollary 5.4(2) and in Section 7.3, even though Drinfeld modules lack canonical auto-duality.","fun_headline_variants_meta":{"raw":{"variants":["Higher Hida theory for Drinfeld modular curves, with Serre duality","New degree-one Hida theory on Drinfeld modular curves","Drinfeld modular curves: higher Hida theory interpolates Serre duality","Higher Hida theory for Drinfeld curves now includes Serre duality","Condensed mathematics enables higher Hida theory on Drinfeld curves"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000656,"raw_usage":{"total_tokens":2941,"prompt_tokens":819,"completion_tokens":2122,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":435,"completion_tokens_details":{"reasoning_tokens":2027}},"tokens_in":435,"tokens_out":2122,"duration_ms":17127,"temperature":1.0,"reasoning_tokens":2027,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:42:09.290298+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Look for a rank-two ordinary Drinfeld module over $A_p$ for which the Hodge–Tate–Taguchi map $HTT: (H^{\\mathrm{can}})^D \\to \\omega$ is not an isomorphism or has a non-trivial zero; the existence of even one such fibre would make $\\omega\\cong\\omega^D$ fail on $X^{\\mathrm{ord}}$, invalidating Corollary 5.4(2) and the identification $D(\\omega^{\\kappa^{\\mathrm{un}}})\\cong\\omega^{\\kappa^{\\mathrm{un}}}\\otimes_{\\Lambda,d}\\Lambda$ that produces the perfect pairing.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the blueprint for higher Hida theory: locally finite operators, projectors, and the pairing argument on the modular curve."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the canonical subgroups, the Hodge–Tate–Taguchi map, and the dual Drinfeld module bundle $\\omega^D$."},{"cited_title":"Taguchi,A duality for finite t-modules, J","cited_arxiv_id":null,"evidence_quote":"Supplies Cartier–Taguchi duality and the pairing with the Carlitz module used throughout the paper."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the degree-zero Hida theory for Drinfeld modular forms that this paper extends, including the congruence $U_p\\equiv T_p$ modulo $\\varpi$."},{"cited_title":"Journal of Number Theory Volume 232, March 2022, Pages 75-100 4, 5, 10, 15","cited_arxiv_id":null,"evidence_quote":"Supplies the level structure, compactification, and canonical subgroup results for the Drinfeld modular curve."},{"cited_title":"Gekeler,De Rham cohomology and the Gauss-Manin connection for Drinfeld modules, p-adic analysis, Lecture Notes in Mathematics 1454 (Springer, Berlin, 1990) 223–255","cited_arxiv_id":null,"evidence_quote":"Supplies the Kodaira–Spencer isomorphism used to compute Serre duality and the dual of $T_p$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the six-operations formalism and the identification of compact-support cohomology used to define $H^1_c$."},{"cited_title":"3, 14, 20","cited_arxiv_id":null,"evidence_quote":"Supplies the condensed-mathematics six-functor formalism in which $H^1_c(X^{\\mathrm{ord}},\\omega^{\\kappa^{\\mathrm{un}}})$ is defined."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Serre–Tate local moduli argument used in Section 3 to prove the function-field Serre–Tate coordinates theorem."}],"review_version":1}