Pith. sign in

REVIEW 3 major objections 4 minor 38 references

Finite Langlands correspondence

T0 review · 3 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper constructs the finite Langlands correspondence: every irreducible ℓ-adic representation of a connected reductive group over a finite field is attached to a special Frobenius-semisimple Langlands parameter, and the preimage of eac

desk verdict The paper builds a categorical finite Langlands correspondence, but the proof of Theorem 5.3 never establishes surjectivity of the L-map, so the main bijection is not fully proved as written. read the letter →

arxiv 2508.15101 v1 pith:5UC4QCHI submitted 2025-08-20 math.NT math.RT

Naoki Imai This is my paper
classification math.NTmath.RT MSC 11F7020C3322E50
keywords finiteLanglandscorrespondenceparametersoverfieldsreductivegroupscategoricalequivalenceunipotentrepresentationsequivariantsheavesLusztigserieslocal
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper's goal is to construct the Langlands correspondence for every connected reductive group over a finite field, not just for the general linear group. It assigns to each irreducible ℓ-adic representation of G(k) a special Frobenius-semisimple Langlands parameter, and it shows that the set of representations over a fixed parameter is naturally bijective to the irreducible representations of an associated finite component group A_φ. The proof works by first establishing a stronger categorical statement: the whole category of representations of G(k) decomposes into blocks of equivariant sheaves on finite groups, one block for each semisimple and unipotent parameter. This categorical rigidity makes the parametrization canonical and extends it to groups with disconnected center. If correct, the result gives a uniform finite-field analogue of the local Langlands correspondence and a finite shadow of the categorical local correspondence.

What carries the argument

The load-bearing mechanism is the categorical equivalence of Theorem 4.5, which decomposes the entire representation category of G(k) into blocks indexed by semisimple and unipotent parameters. Each block is the category of G_c-equivariant sheaves on a finite group G_c with a τ_β-twisted conjugation action and an additional Ω_{c,β}-equivariant structure. From this block category one reads off both the Langlands parameter φ and the finite component group A_φ whose irreducible representations index the fiber over φ. A Whittaker datum is fixed to make the equivalence and the resulting map natural.

What would settle it

Take a small reductive group over F_q with disconnected center, such as SL_2 or a quotient with adjoint center, enumerate the special Frobenius-semisimple Langlands parameters φ and the irreducible ℓ-adic representations of G(k), and check that the fiber over every φ has size |Irr(A_φ)|. A single mismatch would falsify Theorem 5.3; a more structural check is to compute the block category on the right side of Theorem 4.5 for such a group and see whether its Grothendieck group matches the representation ring.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 5.3: after fixing a Whittaker datum, there is a natural map L_G from the irreducible ℓ-adic representations of G(k) to the set of equivalence classes of special Frobenius-semisimple Langlands parameters of Weil–Deligne type, and for each parameter φ the fiber L_G^{-1}(φ) is naturally in bijection with the irreducible representations of A_φ, the finite component group attached to φ. The correspondence is built from Theorem 4.5, a categorical equivalence identifying Rep(G(k)) with a direct sum of categories of equivariant sheaves on finite groups G_c with twisted conjugation. Unipotent representations of connected groups are described first (Proposition 4.1) as

Load-bearing premise

The whole construction depends on a prior categorical endoscopy equivalence behaving naturally and functorially when a finite component group acts on the decomposition; if that compatibility fails, the direct-sum decomposition at the center of the argument does not hold.

Editorial extensions

If this is right

  • Every irreducible ℓ-adic representation of a connected reductive group over a finite field is assigned a well-defined special Frobenius-semisimple Langlands parameter.
  • The fiber over a fixed parameter φ is exactly the set of irreducible representations of the finite component group A_φ, giving a clean description of L-packets.
  • The categorical decomposition rigidifies the parametrization, so groups with disconnected center receive a canonical Lusztig-style parametrization rather than a merely set-theoretic one.
  • The general theorem encompasses the earlier GL_n and SL_n finite Langlands correspondences as special cases.
  • Section 6 conjectures that the finite correspondence controls the tame categorical local Langlands correspondence: the coherent sheaf attached to a representation should be supported on a specified closed set and restrict to a vector bundle built from the representation of A_φ.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper leaves implicit that the naturality of the map L_G should make it compatible with central character twisting and with base change or restriction of scalars; verifying such functoriality is a natural next step.
  • Because the proof is categorical, the correspondence should come with functoriality under morphisms of reductive groups once the underlying endoscopy equivalences are known to be functorial; the paper does not prove such functoriality.
  • A concrete testable extension is to compute the block categories in Theorem 4.5 for a small disconnected-center group, such as a non-split orthogonal group over F_q, and compare the resulting fibers with the character table of G(k), which would exercise the Ω_{c,β}-equivariant structure rather than just the underlying set bijection.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper constructs a Langlands correspondence for connected reductive groups over finite fields. The main categorical result, Theorem 4.5, describes the category of finite-dimensional ℓ-adic representations of G(k) as a direct sum of equivariant sheaf categories associated to finite groups, building on Lusztig's theory and on endoscopic equivalences from [LY21] and [Sol25]. Theorem 5.3 then defines a natural map L_G from Irr_{Qℓ}(G(k)) to the set of special Frobenius-semisimple Weil–Deligne L-parameters, and claims that for every such parameter φ the fiber L_G^{-1}(φ) is in natural bijection with Irr_{Qℓ}(A_φ). The paper also gives a Lusztig-style parametrization of irreducible representations (Theorem 4.12) and formulates a conjecture relating the finite correspondence to the tame categorical local Langlands correspondence.

Significance. If the main theorem is correct, this is a substantial contribution: it provides a canonical, categorical construction of the finite Langlands correspondence for all connected reductive groups over finite fields, with L-packets exactly indexed by irreducible representations of the component group A_φ. The categorical formulation is a genuine strength, as it yields functoriality that a mere bijection would not supply, and it allows the treatment of disconnected centers with a rigidification by a Whittaker datum. The paper also extends Macdonald's GL_n result to a general setting and connects the construction with the categorical local Langlands program. The proof relies heavily on several recent preprints, but the internal logic is mostly coherent; the main missing piece is a surjectivity argument for the constructed map.

major comments (3)
  1. [Theorem 5.3 and its proof] The proof of Theorem 5.3 constructs a map LΨ_G : Irr_{Qℓ}(G(k)) → Ψ_{Qℓ}(G)^sp and shows, for ψ in the image, that the fiber (LΨ_G)^{-1}(ψ) is parametrized by Irr_{Qℓ}(A_ψ), using (5.3) and (5.4). It never proves that LΨ_G is surjective onto Ψ_{Qℓ}(G)^sp. Since A_ψ is a finite group and Irr_{Qℓ}(A_ψ) is nonempty, the asserted bijection for every special φ is equivalent to surjectivity. No counting argument or comparison with the indexing set in Theorem 4.12 is supplied. This is a load-bearing gap in the proof of the main theorem.
  2. [Section 4, after Lemma 4.10 / Theorem 4.12] Theorem 4.12, which is the only result in the paper that could provide a counting or parametrization argument for surjectivity, is proved under the explicit assumption that p is a good prime, made just before Definition 4.11. Theorem 5.3 is stated without this hypothesis, and its proof does not invoke Theorem 4.12. Consequently, even if a surjectivity argument were added using Theorem 4.12, it would cover only good primes. The paper should either extend the argument to all primes or state Theorem 5.3 under the good-prime assumption.
  3. [Theorem 4.5, proof] The proof of Theorem 4.5, and hence of the main correspondence, depends on [LY21, 12.7 Corollary] and its Whittaker-datum rigidification ([LY21, 5.11]), as well as on [Sol25, (3.3),(3.4)]. These are very recent preprints, and the manuscript does not indicate which parts of these inputs are already published and which are still under review. This is not an internal inconsistency, but it is a verification risk: if any of these equivalences fails to be natural under the actions used in equation (4.2), the direct-sum decomposition of Rep(G(k)) and the construction of L_G would lack a foundation. The authors should state the dependence explicitly and, where possible, point to published versions or provide more detail.
minor comments (4)
  1. [Throughout] There are numerous typographical errors: 'parametrizatin', 'run thorough', 'maxmal', 'generaic', 'Joradan decomposition', 'Frobenis', 'automorphsim', 'minimum positive integern'. These should be corrected in a revision.
  2. [Section 5, opening] Theorem 5.3 is stated for 'a reductive algebraic group G over k', but Section 3 defines Langlands parameters only for connected reductive groups, and the Whittaker datum in Section 2 is defined for connected quasi-split groups. Please clarify whether G is assumed connected, or explain how the definitions extend to the non-connected case.
  3. [Lemma 4.10] The first line of the proof, 'It suffices to show Z_AH(uc)(ghβ ˙wβσq)/Z_Gc(gτβ) ≅ Ω_{c,β}', is slightly imprecise: what is needed is a natural isomorphism of the relevant groups, not merely an isomorphism of quotients. The displayed formula could be misread as defining the quotient. Please rephrase.
  4. [Section 3, Definition 3.3] The notation A(φ0) is used both for the quotient π0(Z_{Ĝ}(φ0)/Z(Ĝ)) and for its further quotient by the kernel from the canonical quotient A_{Z_{Ĝ}(φ(I_k))^∘}(φ(G_a)). This overload is potentially confusing; a different symbol for the second quotient would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central categorical equivalence is proved from independent inputs; self-citations are definitional/technical only.

full rationale

The paper's central claim, Theorem 5.3, is derived from Theorem 4.5, whose proof rests on Proposition 4.1 (using Lusztig's categorical center and results of Bezrukavnikov–Finkelberg–Ostrik, Bruguieres–Virelizier, EGNO) and on endoscopic equivalences from Lusztig–Yun [LY21] and Solleveld [Sol25]. These are external or independent prior results, not the finite Langlands correspondence itself. The construction of L_G in Theorem 5.3 translates the categorical data of Theorem 4.5 into a Langlands parameter via (5.3)–(5.4) and Lemma 4.10; the packet parametrization by Irr(A_phi) is a direct consequence of the categorical structure, not a fitted parameter renamed as a prediction. Self-citations to [IV25], [BMIY24], and [Ima24] supply the definition of Langlands parameters for finite fields and auxiliary bijections between SL2-type and Weil–Deligne parameters; these are definitional or technical inputs, and no argument reduces to an unverified self-citation. The paper's note that [LY20] must be replaced by the corrected [LY21] is a correction, not a circularity. The proof's lack of an explicit surjectivity argument for L_G onto Phi^sp is a completeness concern, not a circularity. Overall, no circular step is identifiable.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No fitted constants or invented entities appear. The central claim rests on a tower of external theorems: Lusztig's character classification, Lusztig-Yun categorical endoscopy, Bruguières-Virelizier center computations, Solleveld's disconnected-group endoscopy, and the SL2-to-Weil-Deligne parameter comparison. A Whittaker datum is fixed to rigidify sheaves, and an auxiliary isomorphism (4.8) is chosen.

assumptions (5)
  • domain assumption The unipotent representation category of a connected reductive group is equivalent to the sigma-twisted center of the monoidal cell category C_c^{B2}, and C_c^{B2} is equivalent to the G_c-equivariant functor category on X_c with twist omega.
    Used verbatim in Proposition 4.1 via [Lus15, 6.3(a)], [BFO09, Theorem 4], [BO04, 5.1 Remark], and [Ost14, Remark 2.20].
  • domain assumption The endoscopic categorical equivalence [LY21, 12.7 Corollary] is valid and functorial enough to identify the pi_0(G)-fixed part with the beta-direct summand in Theorem 4.5.
    Invoked in the proof of Theorem 4.5: 'By [LY21, 12.7 Corollary], (4.2) is equivalent to ...'. A Whittaker datum is used to rigidify IC sheaves in [LY21, 5.11].
  • domain assumption Solleveld's endoscopy for representations of disconnected reductive groups over finite fields, [Sol25, (3.3),(3.4)], is valid.
    Used in the proof of Theorem 4.5 for Rep_o(G(k)) isomorphic to Rep_o(G^circ(k))^{pi_0(G)}.
  • domain assumption The bijection between SL2-type and Weil-Deligne Langlands parameters, [BMIY24, Theorem 6.16] and [Ima24, Proposition 1.7], is valid in the ell-adic setting.
    Used in Proposition 5.2 to identify Psi_Qell(G) with Phi_Qell(G), and in the definition of L_G.
  • domain assumption McNinch's results on nilpotent sections and centralizers apply to the relevant groups, with p assumed good for the corresponding part of Section 4.
    Used in Lemma 4.7 and Lemma 4.8, which support Theorem 4.12 under the in-section good prime assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Finite Langlands correspondence." pith.science (2026). https://pith.science/paper/5UC4QCHI

@misc{pith2026250815101,
  author       = {Pith},
  title        = {Pith review of: Finite Langlands correspondence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5UC4QCHI}},
  note         = {Machine review of arXiv:2508.15101}
}
read the original abstract

We construct the Langlands correspondence for connected reductive groups over finite fields, which we call the finite Langlands correspondence. We discuss also its relation with the categorical local Langlands correspondence.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

38 extracted references · 33 canonical work pages

  1. [1]

    On the Macdonald correspondence

    A.-M. Aubert, On the Macdonald correspondence, 2025, arXiv:2501.02332

    work page Pith review arXiv 2025

  2. [2]

    Bezrukavnikov, M

    R. Bezrukavnikov, M. Finkelberg and V. Ostrik, On tensor categories attached to cells in affine W eyl groups. III , Israel J. Math. 170 (2009), 207--234

    work page 2009

  3. [3]

    Bezrukavnikov, M

    R. Bezrukavnikov, M. Finkelberg and V. Ostrik, Character D -modules via D rinfeld center of H arish- C handra bimodules, Invent. Math. 188 (2012), no. 3, 589--620

    work page 2012

  4. [4]

    C. J. Bushnell and G. Henniart, On the derived subgroups of certain unipotent subgroups of reductive groups over infinite fields, Transform. Groups 7 (2002), no. 3, 211--230

    work page 2002

  5. [5]

    C. J. Bushnell and G. Henniart, The local L anglands conjecture for GL(2) , vol. 335 of Grundlehren der Mathematischen Wissenschaften, Springer-Verlag, Berlin, 2006

    work page 2006

  6. [6]

    Bertoloni Meli, N

    A. Bertoloni Meli, N. Imai and A. Youcis, The J acobson-- M orozov M orphism for L anglands P arameters in the R elative S etting, Int. Math. Res. Not. IMRN (2024), no. 6, 5100--5165

    work page 2024

  7. [7]

    Bezrukavnikov and V

    R. Bezrukavnikov and V. Ostrik, On tensor categories attached to cells in affine W eyl groups. II , in Representation theory of algebraic groups and quantum groups, vol. 40 of Adv. Stud. Pure Math., pp. 101--119, Math. Soc. Japan, Tokyo, 2004

    work page 2004

  8. [8]

    Brugui\`eres and A

    A. Brugui\`eres and A. Virelizier, Quantum double of H opf monads and categorical centers, Trans. Amer. Math. Soc. 364 (2012), no. 3, 1225--1279

    work page 2012

Show all 38 references
  1. [9]

    Ben-Zvi, H

    D. Ben-Zvi, H. Chen, D. Helm and D. Nadler, Coherent S pringer theory and the categorical D eligne- L anglands correspondence, Invent. Math. 235 (2024), no. 2, 255--344

    2024

  2. [10]

    Collacciani, A Reduction over finite fields of the tame local Langlands correspondence for SL_n , 2025, arXiv:2501.09085

    E. Collacciani, A Reduction over finite fields of the tame local Langlands correspondence for SL_n , 2025, arXiv:2501.09085

    2025 arXiv

  3. [11]

    J.-F. Dat, D. Helm, R. Kurinczuk and G. Moss, Moduli of L anglands parameters, J. Eur. Math. Soc. (JEMS) 27 (2025), no. 5, 1827--1927

    2025

  4. [12]

    Digne and J

    F. Digne and J. Michel, On L usztig's parametrization of characters of finite groups of L ie type, Ast\' e risque (1990), no. 181-182, 6, 113--156

    1990

  5. [13]

    Digne and J

    F. Digne and J. Michel, Representations of finite groups of L ie type, vol. 95 of London Mathematical Society Student Texts, Cambridge University Press, Cambridge, second edn., 2020

    2020

  6. [14]

    DeBacker and M

    S. DeBacker and M. Reeder, On some generic very cuspidal representations, Compos. Math. 146 (2010), no. 4, 1029--1055

    2010

  7. [15]

    Etingof, S

    P. Etingof, S. Gelaki, D. Nikshych and V. Ostrik, Tensor categories, vol. 205 of Mathematical Surveys and Monographs, American Mathematical Society, Providence, RI, 2015

    2015

  8. [16]

    Fargues and P

    L. Fargues and P. Scholze, Geometrization of the local L anglands correspondence, 2021, arXiv:2102.13459

    2021 arXiv

  9. [17]

    Hellmann, On the derived category of the I wahori- H ecke algebra, Compos

    E. Hellmann, On the derived category of the I wahori- H ecke algebra, Compos. Math. 159 (2023), no. 5, 1042--1110

    2023

  10. [18]

    R. B. Howlett, On the degrees of S teinberg characters of C hevalley groups, Math. Z. 135 (1973/74), 125--135

    1973

  11. [19]

    Imai, Local L anglands correspondences in -adic coefficients, Manuscripta Math

    N. Imai, Local L anglands correspondences in -adic coefficients, Manuscripta Math. 175 (2024), no. 1-2, 345--364

    2024

  12. [20]

    Imai and D

    N. Imai and D. A. Vogan, Jr., Langlands parameters for reductive groups over finite fields, 2025, arXiv:2506.06961

    2025 arXiv

  13. [21]

    Kaletha, On the local Langlands conjectures for disconnected groups, 2022, arXiv:2210.02519

    T. Kaletha, On the local Langlands conjectures for disconnected groups, 2022, arXiv:2210.02519

    2022 arXiv

  14. [22]

    Kerler and V

    T. Kerler and V. V. Lyubashenko, Non-semisimple topological quantum field theories for 3-manifolds with corners, vol. 1765 of Lecture Notes in Mathematics, Springer-Verlag, Berlin, 2001

    2001

  15. [23]

    Lusztig, Characters of reductive groups over a finite field, vol

    G. Lusztig, Characters of reductive groups over a finite field, vol. 107 of Annals of Mathematics Studies, Princeton University Press, Princeton, NJ, 1984

    1984

  16. [24]

    Lusztig, Leading coefficients of character values of H ecke algebras, in The A rcata C onference on R epresentations of F inite G roups ( A rcata, C alif., 1986), vol

    G. Lusztig, Leading coefficients of character values of H ecke algebras, in The A rcata C onference on R epresentations of F inite G roups ( A rcata, C alif., 1986), vol. 47, Part 2 of Proc. Sympos. Pure Math., pp. 235--262, Amer. Math. Soc., Providence, RI, 1987

    1986

  17. [25]

    Lusztig, On the representations of reductive groups with disconnected centre, 168, pp

    G. Lusztig, On the representations of reductive groups with disconnected centre, 168, pp. 10, 157--166, 1988, orbites unipotentes et repr\' e sentations, I

    1988

  18. [26]

    Lusztig, Irreducible representations of finite spin groups, Represent

    G. Lusztig, Irreducible representations of finite spin groups, Represent. Theory 12 (2008), 1--36

    2008

  19. [27]

    Lusztig, Families and S pringer's correspondence, Pacific J

    G. Lusztig, Families and S pringer's correspondence, Pacific J. Math. 267 (2014), no. 2, 431--450

    2014

  20. [28]

    Lusztig, Unipotent representations as a categorical centre, Represent

    G. Lusztig, Unipotent representations as a categorical centre, Represent. Theory 19 (2015), 211--235

    2015

  21. [29]

    Lusztig and Z

    G. Lusztig and Z. Yun, Endoscopy for H ecke categories, character sheaves and representations, Forum Math. Pi 8 (2020), e12, 93

    2020

  22. [30]

    Lusztig and Z

    G. Lusztig and Z. Yun, Endoscopy for H ecke categories, character sheaves and representations, 2021, arXiv:1904.01176v3

    2021 arXiv

  23. [31]

    I. G. Macdonald, Zeta functions attached to finite general linear groups, Math. Ann. 249 (1980), no. 1, 1--15

    1980

  24. [32]

    G. J. McNinch, The centralizer of a nilpotent section, Nagoya Math. J. 190 (2008), 129--181

    2008

  25. [33]

    Ostrik, Tensor categories attached to exceptional cells in W eyl groups, Int

    V. Ostrik, Tensor categories attached to exceptional cells in W eyl groups, Int. Math. Res. Not. IMRN (2014), no. 16, 4521--4533

    2014

  26. [34]

    III : S tructure des sch\'emas en groupes r\'eductifs, S\'eminaire de G\'eom\'etrie Alg\'ebrique du Bois Marie 1962/64 (SGA 3)

    Sch\'emas en groupes. III : S tructure des sch\'emas en groupes r\'eductifs, S\'eminaire de G\'eom\'etrie Alg\'ebrique du Bois Marie 1962/64 (SGA 3). Dirig\'e par M. Demazure et A. Grothendieck. Lecture Notes in Mathematics, Vol. 153, Springer-Verlag, Berlin-New York, 1970

    1962

  27. [35]

    Solleveld, Endoscopy for representations of disconnected reductive groups over finite fields, 2025, arXiv:2507.16304

    M. Solleveld, Endoscopy for representations of disconnected reductive groups over finite fields, 2025, arXiv:2507.16304

    2025 arXiv

  28. [36]

    Yun, Rigidity in automorphic representations and local systems, in Current developments in mathematics 2013, pp

    Z. Yun, Rigidity in automorphic representations and local systems, in Current developments in mathematics 2013, pp. 73--168, Int. Press, Somerville, MA, 2014

    2013

  29. [37]

    Zhu, Coherent sheaves on the stack of Langlands parameters, 2020, arXiv:2008.02998

    X. Zhu, Coherent sheaves on the stack of Langlands parameters, 2020, arXiv:2008.02998

    2020

  30. [38]

    Zhu, Tame categorical local Langlands correspondence, 2025, arXiv:2504.07482

    X. Zhu, Tame categorical local Langlands correspondence, 2025, arXiv:2504.07482

    2025 arXiv

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.