Pith. sign in

REVIEW 2 major objections 4 minor 13 references

Typical representations for level zero Bernstein components of ${\rm GL}_n(F)$

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read For a level-zero Bernstein component of $\mathrm{GL}_n(F)$, the typical representations are exactly the irreducible subrepresentations of an explicit induced type.

desk verdict A plausible and well-organized extension of the level-zero typical-representation classification, but the nr>1 case relies on an unproved containment in Proposition 4.3 that the referee must force the author to justify. read the letter →

arxiv 1908.03392 v1 pith:CYXK5YWE submitted 2019-08-09 math.RT math.NT

Santosh Nadimpalli This is my paper · ORCID
classification math.RTmath.NT MSC 22E5011S3722E35
keywords typicalrepresentationBernsteincomponentlevelzeroBushnell-KutzkotypesupercuspidalcuspidalsupportparabolicinductionGL_n
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

This paper classifies the typical representations for level-zero Bernstein components of a non-Archimedean local field $F$. A typical representation is an irreducible representation of the maximal compact subgroup $\mathrm{GL}_n(O_F)$ whose appearance inside an irreducible smooth representation $\pi$ forces $\pi$ to have a prescribed cuspidal support. The main theorem states that for a level-zero component $s=[M_I,\sigma_I]$, every typical representation occurs inside the induced representation $\operatorname{ind}_{P_I(1)}^{\mathrm{GL}_n(O_F)}(\tau_I)$ built from the level-one Bushnell--Kutzko type; conversely, every irreducible subrepresentation of that induced representation is typical. The paper also proves that the multiplicity of each typical representation in this induced representation equals its multiplicity in the corresponding parabolic induction, and the result holds independently of the characteristic of the base field.

What carries the argument

The carrying object is the induced representation $\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I)$ and its filtration by $\operatorname{ind}_{P_I(m)}^{K_n}(\tau_I)$, where $P_I(m)$ is the preimage of $P_I(O_F/P_F^m)$ in $K_n$. Iwahori decomposition lets $P_I(m)\cap U_I$ and $P_I(m)\cap \overline{U}_I$ act trivially, so Lemma 2.5 expresses the full induced representation as a union of finite-dimensional pieces. The proof decomposes each successive quotient by Clifford theory into the identity character plus non-trivial characters $\eta$, whose stabilizers $Z(\eta)$ are controlled by a matrix-stabilizer lemma over the residue field; Lemma 3.9 then converts a trivial intersection with a unipotent group into the existence of a non-cuspidal representation containing the given piece. In the rank-one case $n_r=1$, Zelevinsky's derivative functors and Casselman's restriction decomposition for $\mathrm{GL}_2$ are used. Together these force all subrepresentations of the complement to be atypical, which is the engine behind the classification.

What would settle it

Take a level-zero component $s=[M_I,\sigma_I]$ for a small $n$, list all irreducible subrepresentations of $\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I)$, and check whether any other irreducible $K_n$-type is typical for $s$ by testing whether it occurs in $\operatorname{res}_{K_n}i_{P_I}^{\mathrm{GL}_n(F)}(\sigma_I)$; finding such a $\Gamma$ outside the listed set, or finding one whose multiplicity in the two sides of Corollary 3.3 differs, would refute the classification.

Watch

Extended reading notes

Core claim

The paper's central claim, Theorem 1.1 and Corollary 3.3, is that for a level-zero inertial class $s=[M_I,\sigma_I]$ with ordered partition $I=(n_1,\dots,n_r)$ of $n$, the $K_n=\mathrm{GL}_n(O_F)$-irreducible subrepresentations of $\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I)$ are precisely the typical representations for $s$. Here $\tau_I=\tau_1\boxtimes\cdots\boxtimes\tau_r$, where each $\tau_i$ is the unique typical representation attached to the supercuspidal component $[G_{n_i},\sigma_i]$, realized as the inflation of a cuspidal representation of $\mathrm{GL}_{n_i}(k_F)$. The proof shows that the complement $U_m(\tau_I)$ in $\operatorname{ind}_{P_I(m)}^{K_n}(\tau_I)$ contains only atypical subrepresentations, so no typical representation can hide outside the level-one induced representation. In addition, for every typical $\Gamma$, the paper proves $\dim_{\mathbb{C}}\operatorname{Hom}_{K_n}(\Gamma,\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I))=\dim_{\mathbb{C}}\operatorname{Hom}_{K_n}(\Gamma,i_{P_I}^{\mathrm{GL}_n(F)}(\sigma_I))$.

Load-bearing premise

The argument leans on the previously proved uniqueness of typical representations for the basic cuspidal pieces, a theorem this paper cites rather than proves; if that uniqueness failed, the complement decomposition driving the induction would break.

Editorial extensions

If this is right

  • For every level-zero Bernstein component $s$, the set of typical representations is finite: it is exactly the finite set of irreducible subrepresentations of $\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I)$.
  • A $K_n$-representation is typical for $s$ exactly when it appears in this single explicitly given induced representation, so inertial-support detection for level-zero components reduces to listing that finite set.
  • The multiplicity identity means the numerical data carried by typical representations over $K_n$ coincide with embeddings of $\sigma_I$ in parabolic induction, allowing computations on the compact group to be translated into parabolic-induction computations.
  • Because the result is independent of the characteristic of $F$, it covers both characteristic-zero local fields and equal-characteristic local fields such as Laurent series fields over finite fields.

Reading between the lines

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

  • Editor's inference: the explicit finite list of typical representations should make the level-zero inertial-support detection problem effectively computable for small $n$, for example by tabulating the irreducible subrepresentations of $\operatorname{ind}_{P_I(1)}^{K_n}(\tau_I)$ for each residue cardinality.
  • Editor's inference: if the complement argument could be reworked without invoking the cuspidal unicity theorem, the classification would likely have a natural analogue for positive-level components where typical representations are not known to be unique.
  • Editor's inference: the equality of multiplicities suggests that typical representations may index a natural basis of the space of $K_n$-fixed vectors in parabolic induction, possibly visible through a Hecke-module structure attached to the type.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper classifies typical representations for level-zero Bernstein components of GL_n(F). Theorem 1.1 and Corollary 3.3 assert that for a level-zero component s=[M_I,σ_I], the K_n-irreducible subrepresentations of ind_{P_I(1)}^{K_n}(τ_I) are precisely the typical representations for s, and that every typical representation occurs there with the same multiplicity as in the parabolic induction i_{P_I}^{GL_n(F)}(σ_I). The proof combines Bushnell-Kutzko types, Paškūnas' unicity theorem for supercuspidal components, the Iwasawa decomposition to restrict parabolic induction to K_n, a filtration by the compact subgroups P_I(m), and an induction on n; the cases n_r=1 (Borel) and n_r>1 are treated separately, with Clifford theory and finite-field lemmas in the latter.

Significance. If the main theorem is correct, the paper completes the level-zero classification of typical representations, extends the Henniart and Paškūnas results for cuspidal components to all level-zero Bernstein components, and provides a multiplicity formula. The strategy is well chosen and the paper is honest about relying on external inputs: Bushnell-Kutzko types, Paškūnas' unicity, Casselman's restriction theorem, and Zelevinsky's derivative formalism are cited explicitly, and I see no fitted parameters or circularity. However, two steps in the n_r>1 branch of the proof are not justified as written; both are load-bearing for Theorem 3.2 and Corollary 3.3, so the main claim is not yet established in the present text.

major comments (2)
  1. [4, Proposition 4.3, Eq. (22)] The proof of Proposition 4.3 asserts that the representation in Eq. (22), namely ind_{Z(η_k)}^{K_{n+1}}(U_{η_k} ⊗ res_{Z∩M_I} τ'_I), 'occurs as a subrepresentation of' ind_{P_I∩K_{n+1}}^{K_{n+1}}(τ'_I). This containment is the only bridge from the Clifford-theoretic decomposition to parabolic induction, and it is not a formal consequence of transitivity of induction or Frobenius reciprocity. The subgroup Z(η_k) contains K_I(m)∩\bar U, on which the Clifford character U_{η_k} is nontrivial, while the target representation is trivial on the lower unipotent part, and Z(η_k) is not contained in P_I∩K_{n+1}. An embedding therefore requires an explicit Mackey double-coset computation over Z(η_k)\K_{n+1}/(P_I∩K_{n+1}), or an equivalent argument, and none is supplied. Since Proposition 4.3 is the decisive step for all level-zero components with r>1 and n_r>1, Theorem 3.2 and Corollary 3.3 are unsupported for that case until this containment is proved.
  2. [3.1, Lemma 3.8] In the proof of Lemma 3.8, the passage from Eq. (9) to the equation involving M_{ll} drops the off-diagonal terms M_{lj}U_j^{tr} for j<l. Since M is block upper triangular, the l-th block of Eq. (9) is Σ_{j≤l} M_{lj}U_j^{tr} = U_l^{tr}B, and these lower-block terms need not vanish even when l is chosen maximal with U_l≠0. Consequently the claim that M_{ll} preserves ker T1 (or that B preserves ker T2) is not established as written. This lemma is used in Proposition 4.3 to produce the unipotent subgroup U with H∩U={id}, so the gap affects the same load-bearing step of the main theorem.
minor comments (4)
  1. [2.1] In the definition of U_I(R), 'unipotent unipotent matrices' should read 'unipotent matrices'.
  2. [3, Lemma 3.9] In the proof of Lemma 3.9, the symbol γ in ind_H^G(γ) is undefined; it should be the representation ξ, and the intended Mackey-decomposition argument should be spelled out.
  3. [4, n_r=1 case] The notation for the last character wavers between χ_n and χ_{n+1}; for example the summand U_m(χ_{I_n})⊠χ_n should presumably be U_m(χ_{I_n})⊠χ_{n+1}.
  4. [4, Proposition 4.3] Eq. (22) is referred to in the text as 'The representation 22'; the equation number should be typeset consistently.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the derivation is self-contained against external theorems and no fitted parameter or self-citation chain reduces the main claim to its inputs.

full rationale

The paper's derivation chain is not circular. Theorem 1.1 and Corollary 3.3 assert that typical representations for level-zero Bernstein components occur as subrepresentations of the Bushnell-Kutzko induced type, and the proof proceeds by induction using explicit decompositions of induced representations. The load-bearing external inputs are independent established results: Bushnell-Kutzko types ([BK99],[BK93]), Paskunas' unicity theorem for supercuspidal typical representations ([Pas05, Theorem 8.1]), Casselman's restriction theorem for GL_2, and Zelevinsky's derivative functors. None of these is equivalent to the paper's target statement, and none is fitted to the data being predicted. The paper cites Paskunas' theorem rather than proving it, but that theorem concerns a different, supercuspidal case and is parameter-free with assumptions that do not include the conclusion of this paper; it is therefore genuine external support, not a self-citation chain. The only notable concern is in Proposition 4.3, where the assertion that the representation in equation (22) occurs as a subrepresentation of ind_{P_I ∩ K_{n+1}}^{K_{n+1}}(τ'_I) is stated without an explicit Mackey double-coset argument. That is a possible gap in the proof, but it is not circularity: the containment is not claimed to follow from the theorem being proved, and no input is renamed as a prediction. No parameter is fitted, no quantity is defined in terms of the target result, and no self-citation is load-bearing. Accordingly, the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

Pure mathematics paper; no parameters fitted to data and no new postulated objects. The proof depends on established structural theorems, including Bushnell-Kutzko, Paskunas, Casselman, and Zelevinsky, plus standard representation theory. The main new content is the inductive argument in Section 4.

assumptions (8)
  • domain assumption Bushnell-Kutzko semisimple type for a level-zero inertial class s=[M_I,σ_I] is the pair (P_I(1),τ_I), from [BK99, Section 8.3.1].
    Invoked in Section 3 before Definition 3.1 and in Corollary 3.3 to convert nonvanishing Hom_{P_I(1)}(τ_I,π) into π ∈ A_n(s). This is the bridge between the compact-open induced representation and the Bernstein component.
  • domain assumption Paskunas' unicity theorem: for a supercuspidal component [G_n,σ], there is a unique typical representation, namely the induced type, and it occurs with multiplicity one in unramified twists (Theorem 2.2, [Pas05, Theorem 8.1]).
    Used in Lemma 2.4 to split res_{K_{n_i}}(σ_i) as τ_i ⊕ τ_i^1 with τ_i^1 atypical; this split is load-bearing for the induction in the main theorem.
  • standard math Bernstein decomposition of the category of smooth representations of GL_n(F) into blocks indexed by inertial classes, from [Ren10, VI.7.2].
    Gives the definition of A_n(s), inertial support, and the direct-product decomposition used throughout the paper.
  • standard math Zelevinsky's derivative functors, the filtration on mirabolic representations, and the ring structure on R=⊕R_n for finite GL_n(k_F), from [Zel81, Chapter 3, §13].
    Used in the nr=1 case to compute D(ind_{B_n(k_F)}^{GL_n(k_F)}(χ_I)) and to identify the pieces (Φ_+)^{k-1}Ψ_+(X_{n-k}) inside the restriction to the mirabolic subgroup.
  • domain assumption Casselman's restriction formula for irreducible representations of GL_2(F) to GL_2(O_F), from [Cas73b, Props. 1-2 and Thm. 1].
    Used in the nr=1 case to show that the remaining representation (17) is contained in a parabolic induction with a level-zero supercuspidal GL2 piece, making its irreducible subrepresentations atypical.
  • standard math Derivative computation for supercuspidal representations of GL_n(k_F): π^(n)=1 and π^(k)=0 for k<n, from [Zel81, Ch. 3, §13] after Gelfand-Kazhdan.
    Used together with the ring homomorphism D to expand D(ind χ_I) as a product of (χ_i+1_R).
  • domain assumption A cuspidal representation of a finite reductive group has no nonzero vectors fixed by the unipotent radical of a proper parabolic subgroup.
    Used in Lemma 3.9, attributed to Paskunas, to prove that any irreducible ξ of H with H∩U={id} embeds into some non-cuspidal representation.
  • standard math Standard Frobenius reciprocity, Mackey decomposition, Clifford theory, and Iwahori decompositions for congruence subgroups.
    Used throughout Sections 2.2, 3.1, and 4 to pass between restrictions and inductions; specifically in Lemmas 2.5, 2.6, 3.5, 3.6, and equation (8).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Typical representations for level zero Bernstein components of ${\rm GL}_n(F)$." pith.science (2026). https://pith.science/paper/CYXK5YWE

@misc{pith2026190803392,
  author       = {Pith},
  title        = {Pith review of: Typical representations for level zero Bernstein components of $\rm GL_n(F)$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CYXK5YWE}},
  note         = {Machine review of arXiv:1908.03392}
}
abstract

Let $F$ be a non-discrete non-Archimedean locally compact field. In this article for a level zero Bernstein component $s$, we classify those irreducible smooth representations of ${\rm GL}_n{\integers{F}}$ (called typical representations) whose appearance in a smooth irreducible representation $\pi$ of ${\rm GL}_n{F}$ implies that the cuspidal support of $\pi$ is $s$. These results extend, for level zero representations, the results of Henniart and Pa\v{s}k\={u}nas on cuspidal representations. The results are independent of the characteristic of the base field.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

13 extracted references · 11 canonical work pages

  1. [1]

    Bushnell and Guy Henniart, The local L anglands conjecture for GL(2) , Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol

    Colin J. Bushnell and Guy Henniart, The local L anglands conjecture for GL(2) , Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol. 335, Springer-Verlag, Berlin, 2006. 2234120 (2007m:22013)

    work page 2006

  2. [2]

    Bushnell and Philip C

    Colin J. Bushnell and Philip C. Kutzko, The admissible dual of GL (N) via compact open subgroups , Annals of Mathematics Studies, vol. 129, Princeton University Press, Princeton, NJ, 1993. 1204652 (94h:22007)

    work page 1993

  3. [3]

    119 (1999), no

    , Semisimple types in GL _n , Compositio Math. 119 (1999), no. 1, 53--97. 1711578 (2000i:20072)

    work page 1999

  4. [4]

    Christophe Breuil and Ariane M \'e zard, Multiplicit\'es modulaires et repr\'esentations de GL _2( Z _p) et de Gal ( Q _p/ Q _p) en l=p , Duke Math. J. 115 (2002), no. 2, 205--310, With an appendix by Guy Henniart. 1944572 (2004i:11052)

    work page 2002

  5. [5]

    I. N. Bern s te n and A. V. Zelevinski , Representations of the group GL(n,F), where F is a local non- A rchimedean field , Uspehi Mat. Nauk 31 (1976), no. 3(189), 5--70. 0425030 (54 \#12988)

    1976

  6. [6]

    William Casselman, On some results of A tkin and L ehner , Math. Ann. 201 (1973), 301--314. 0337789

    work page 1973

  7. [7]

    , The restriction of a representation of GL _ 2 (k) to GL _ 2 ( o ) , Math. Ann. 206 (1973), 311--318. 0338274 (49 \#3040)

    work page 1973

  8. [8]

    Matthew Emerton and Toby Gee, A geometric perspective on the B reuil- M \'ezard conjecture , J. Inst. Math. Jussieu 13 (2014), no. 1, 183--223. 3134019

    2014

Show all 13 references
  1. [9]

    London Math

    Vytautas Paskunas, Unicity of types for supercuspidal representations of GL _N , Proc. London Math. Soc. (3) 91 (2005), no. 3, 623--654. 2180458 (2007b:22018)

    2005

  2. [10]

    17, Soci\'et\'e Math\'ematique de France, Paris, 2010

    David Renard, Repr\'esentations des groupes r\'eductifs p -adiques , Cours Sp\'ecialis\'es [Specialized Courses], vol. 17, Soci\'et\'e Math\'ematique de France, Paris, 2010. 2567785 (2011d:22019)

    2010

  3. [11]

    Schneider and E.-W

    P. Schneider and E.-W. Zink, K -types for the tempered components of a p -adic general linear group , J. Reine Angew. Math. 517 (1999), 161--208, With an appendix by Schneider and U. Stuhler. 1728541 (2001f:22029)

    1999

  4. [12]

    Conley William, Inertial types and automorphic representations with prescribed ramification, Phd Thesis, UCLA (2010)

    2010

  5. [13]

    Zelevinsky, Representations of finite classical groups, Lecture Notes in Mathematics, vol

    Andrey V. Zelevinsky, Representations of finite classical groups, Lecture Notes in Mathematics, vol. 869, Springer-Verlag, Berlin-New York, 1981, A Hopf algebra approach. 643482

    1981

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.