🔢 mathematics

Towards the $p$ -adic Hodge parameters in semistable representations of $\mathrm{GL}_n(\mathrm{Q}_p)$

By synthesizing methods from Ding, Breuil–Ding, and Qian, this paper captures the full $p$ -adic Hodge parameters of non-critical semistable Galois representations through specific Steinberg subquotients and constructs an explicit locally analytic representation that determines these parameters, particularly when the monodromy rank is at most one, thereby advancing the $p$ -adic Langlands program for semistable cases.

Original authors: Yiqin He

Published 2026-04-03

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Yiqin He

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to solve a massive, intricate puzzle. This puzzle represents a hidden mathematical object called a Galois representation (let's call it "The Secret"). This object lives in the world of numbers, specifically involving prime numbers like $p$ .

The paper by Yiqin He is about figuring out how to reconstruct "The Secret" just by looking at its shadow or its reflection in a different world called Automorphic Representations (let's call this "The Mirror").

Here is the breakdown of the paper using simple analogies:

1. The Two Worlds: The Secret and The Mirror

In mathematics, there is a famous conjecture called the p-adic Langlands Program. Think of it as a Rosetta Stone that translates between two languages:

Language A (Galois): Describes symmetries of numbers (The Secret). It has a specific "fingerprint" called Hodge parameters. These parameters are like the unique DNA of the object.
Language B (Automorphic): Describes functions and waves (The Mirror).

The big question is: If we see a specific pattern in the Mirror (Language B), can we perfectly reconstruct the DNA (Hodge parameters) of the Secret (Language A)?

2. The Problem: The "Semistable" Fog

Mathematicians have already solved this puzzle for two extreme cases:

The Crystalline Case: The object is very clear and rigid, like a diamond. We know how to read its DNA from the Mirror.
The Steinberg Case: The object is very fluid and chaotic, like a storm. We also figured out how to read its DNA here.

But there is a middle ground called the Semistable Case. Imagine an object that is part crystal and part storm. It's "semi-stable." In this foggy middle ground, the DNA is harder to read because the "crystal" parts and the "storm" parts are tangled together. Previous methods couldn't fully untangle them to get the full picture.

3. The Solution: The "Lego" Strategy

Yiqin He's paper proposes a clever new way to untangle the mess. Instead of trying to look at the whole tangled object at once, the author suggests breaking it down into smaller, manageable Lego blocks.

The Blocks (Steinberg Subquotients): The author realizes that even in this messy "semistable" object, there are smaller pieces that are pure "storms" (Steinberg blocks).
The Glue (Crystalline Parameters): The way these blocks are glued together holds the missing information.

The Analogy: Imagine you have a complex machine. You can't see the blueprint, but you can see the gears (the blocks) and the springs (the glue) connecting them.

The author says: "If we look at the gears individually, we learn some things. If we look at how the gears connect to each other (the 'crystalline' glue), we learn the rest."
By combining the information from the gears and the glue, we can reconstruct the entire blueprint.

4. The New Tool: "L-Invariants" as a Decoder Ring

To read the information from the "Mirror" side, the author uses a special tool called Breuil-Schraen L-invariants.

Think of these as decoder rings.
The "Mirror" (the automorphic representation) is a complex song. The L-invariants are the specific notes in that song that tell you exactly how the gears in the "Secret" machine are connected.
The paper proves that if you listen to the right combination of these notes (higher extension groups), you can decode the exact DNA of the original object.

5. The Result: A New Map

The paper achieves three main things:

Mapping the DNA: It proves that you can recover all the missing DNA (Hodge parameters) of the "semistable" object by looking at specific parts of the Mirror.
Building a New Machine: It constructs a specific mathematical object (a locally analytic representation) that acts as a perfect "decoder." If you have this object, you automatically know the DNA of the original Secret.
Connecting the Worlds: It shows that this new decoder is actually a piece of a much larger, global puzzle. This gives strong evidence that the "Rosetta Stone" (Langlands Program) works even in these messy, foggy middle cases.

Summary in One Sentence

Yiqin He figured out how to reconstruct a complex, half-crystal/half-storm mathematical object by breaking it into smaller, known pieces and using a new "decoder ring" to read the hidden connections between them, proving that the bridge between the world of numbers and the world of functions is stronger than we thought.

Why does this matter?
It's like finding a new way to read a lost language. Every time we solve a piece of the Langlands puzzle, we get closer to understanding the fundamental laws of how numbers interact, which has deep implications for cryptography, physics, and our understanding of the universe's mathematical structure.

1. Problem Statement

The paper addresses a central problem in the p-adic Langlands program: reconstructing the Hodge filtration (and thus the full p-adic Galois representation) from locally analytic representations of $GL_n(\mathbb{Q}_p)$ .

Context: Let $\rho_p$ be an $n$ -dimensional non-critical semistable p-adic Galois representation of $Gal(\overline{\mathbb{Q}}_p/\mathbb{Q}_p)$ with regular Hodge–Tate weights. By Fontaine's theory, $\rho_p$ is determined by its associated filtered $(\varphi, N)$ -module (or $(\varphi, \Gamma)$ -module $D$ over the Robba ring).
The Gap: While the "locally algebraic" part of the conjectural locally analytic representation $\pi_{an}(\rho_p)$ encodes the Weil–Deligne representation and Hodge–Tate weights, it misses the Hodge filtration data.
Specific Challenge: Previous work (Ding, Breuil–Ding) successfully handled the crystalline case ( $S_0 = \emptyset$ ), and Qian handled the Steinberg case ( $S_0 = \Delta$ ). This paper aims to unify and extend these results to the general semistable case, where the monodromy operator $N$ has an intermediate rank ( $S_0 \subsetneq \Delta$ ). The goal is to capture the full moduli space of p-adic Hodge parameters, denoted $\Phi_{nc, \Delta}(S_0)$ , on the automorphic side.

2. Methodology

The author employs a combination of deformation theory, representation theory of $p$ -adic groups, and geometric methods on flag varieties.

A. Decomposition of $(\varphi, \Gamma)$ -modules

The core strategy involves decomposing the semistable module $D$ into Steinberg subquotients.

For any triangulation $F_u$ associated with a Weyl group element $u \in W_n^{S_0}$ , $D$ admits a parabolic filtration $F_{S_0(u)}$ with graded pieces $\{E_{u,i}\}$ that are Steinberg $(\varphi, \Gamma)$ -modules.
The paper defines a collection of subquotients $CST$ (Steinberg blocks and their "crystalline" interactions) and specific morphisms $\iota_{r,q}^D$ between them.
Key Insight: The full Hodge filtration $L(D)$ $L (D)$ can be recovered by combining:
1. The Hodge parameters within the Steinberg blocks (captured by Breuil–Schraen $L$ -invariants).
2. The "crystalline" Hodge parameters between different Steinberg blocks.
3. The Hodge parameters of the maximal crystalline subquotient $D_{cr}$ .

B. Higher Extension Groups and $L$ -invariants

The paper utilizes the higher extension groups of locally analytic generalized Steinberg representations.

It builds on Qian's work to compute $\text{Ext}^\bullet$ between parabolic inductions of Steinberg representations.
Breuil–Schraen $L$ -invariants: These are defined as codimension-1 subspaces in specific extension groups. The paper proves that these $L$ -invariants encode the Hodge parameters of the Steinberg blocks.
Cup Products: The structure of these extension groups is analyzed via cup products, allowing the reduction of global extension problems to local (Levi) block problems.

C. Construction of Locally Analytic Representations

The author constructs explicit locally analytic representations $\pi_1^\flat(D)$ , $\pi_1^\sharp(D)$ , and $\pi_1^\diamond(D)$ (and a conjectural $\pi_1^+(D)$ ).

These are constructed as amalgamated sums (pushouts) of extensions involving:
- Locally algebraic vectors $\pi_1^{lalg}$ .
- Generalized Steinberg representations.
- Orlik–Strauch functors applied to specific smooth representations.
The representations are designed such that their extension classes in $\text{Ext}^1_G(\pi_1^{lalg}, \pi_1^\star)$ correspond to the Hodge filtration data.

D. Local-Global Compatibility

The paper verifies these constructions in the patched setting (Taylor–Wiles method).

It relates the constructed representations to the patched eigenvariety $X_p(\rho)$ .
It shows that under mild hypotheses, the constructed representation $\pi_{min}^\sharp(D)$ appears as a subrepresentation of the global patched Banach representation $\Pi_\infty$ associated with $\rho_p$ .

3. Key Contributions and Results

Theoretical Characterization of Hodge Parameters

Theorem 1.2 & 1.3: The paper proves that the $(\varphi, \Gamma)$ $(φ, Γ)$ -module $D$ $D$ is uniquely determined by:
- The collection of Steinberg blocks $CST$ (encoding internal Hodge parameters).
- The maximal crystalline subquotient $D_{cr}$ (or equivalently, the morphisms $\iota_{r,q}^D$ between blocks).
Geometric Reformulation: The moduli space $\Phi_{nc, \Delta}(S_0)$ is shown to be injectively mapped into a product of smaller moduli spaces associated with Steinberg blocks and the crystalline case. This is reformulated in terms of Lie algebras (Theorem 3.13), showing that the kernel of a specific map between Lie algebra representations determines the Hodge parameters.

Construction of Automorphic Counterparts

Theorem 6.8 & 6.10: The author constructs explicit locally analytic representations $\pi_1^\sharp(D)$ $π_{1}^{♯} (D)$ and $\pi_1^\diamond(D)$ $π_{1}^{⋄} (D)$ .
- There exist surjective maps $t_D^\sharp$ and $t_D^\diamond$ from the extension group $\text{Ext}^1_G(\pi_1^{lalg}, \pi_1^\sharp)$ to the space of Hodge parameters (specifically, the space of endomorphisms preserving the filtration).
- The kernel of these maps determines the Hodge filtration lines $\{L_i\}$ modulo the center.
Rank 1 Monodromy: If the monodromy rank of $\rho_p$ is at most 1, the constructed representation $\pi_1^\sharp(D)$ uniquely determines the Galois representation $\rho_p$ .

Local-Global Compatibility

Theorem 7.4: In the patched setting, the constructed representation $\pi_{min}^\sharp(D)$ embeds into the global patched representation $\Pi_\infty[\mathfrak{m}_\rho]$ .
Corollary 7.5: If the monodromy rank is $\le 1$ , the global representation $\Pi_\infty[\mathfrak{m}_\rho]$ determines the local Galois representation $D$ . This provides strong evidence for the local-global compatibility conjecture in the semistable case.

Explicit Examples

Appendix: The paper provides explicit diagrams of the representations for $GL_2(\mathbb{Q}_p)$ and $GL_3(\mathbb{Q}_p)$ . For $GL_3$ , it explicitly constructs the representation for the case $S_0 = \{1\}$ (monodromy rank 1), showing how "simple" $L$ -invariants and "higher" $L$ -invariants appear in different "branches" of the representation.

4. Significance

Bridging the Gap: This work is a major step toward the p-adic local Langlands correspondence for $GL_n(\mathbb{Q}_p)$ in the general semistable case, moving beyond the crystalline and pure Steinberg cases.
Unification of Methods: It successfully combines the methods of Ding (crystalline case), Breuil–Ding, and Qian (Steinberg case) into a coherent framework for general semistable representations.
New Evidence for Conjectures: By constructing explicit representations that recover the Hodge filtration and proving their local-global compatibility, the paper provides concrete evidence for the existence of a locally analytic Langlands correspondence that captures all $p$ -adic Hodge theoretic data.
Geometric Insight: The reformulation of the problem in terms of flag varieties and Lie algebras offers a new geometric perspective on how Hodge parameters are distributed across different refinements of the Galois representation.

In summary, Yiqin He demonstrates that the full p-adic Hodge parameters of a semistable representation can be "captured" on the automorphic side by carefully analyzing the extension structures between locally analytic representations associated with the Steinberg blocks of the underlying $(\varphi, \Gamma)$ -module.