🔢 mathematics

Strong Hybrid Subconvexity for Twisted Selfdual $\mathrm{GL}_3$ $L$ -Functions

This paper establishes strong hybrid subconvexity bounds for twisted selfdual $\mathrm{GL}_3$ $L$ -functions and certain $\mathrm{GL}_3 \times \mathrm{GL}_2$ Rankin-Selberg $L$ -functions in both the $q$ and $t$ aspects by utilizing an explicit $\mathrm{GL}_3 \times \mathrm{GL}_2 \leftrightsquigarrow \mathrm{GL}_4 \times \mathrm{GL}_1$ spectral reciprocity formula and a new Lindelöf-on-average bound for Dirichlet $L$ -functions.

Original authors: Soumendra Ganguly, Peter Humphries, Yongxiao Lin, Ramon Nunes

Published 2026-03-17

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Soumendra Ganguly, Peter Humphries, Yongxiao Lin, Ramon Nunes

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 predict the weather in a very chaotic, high-dimensional city. In the world of mathematics, this "weather" is represented by L-functions. These are complex mathematical objects that encode deep secrets about numbers, much like how a weather map encodes secrets about wind and rain.

The paper you are asking about is a major breakthrough in understanding one specific type of L-function (related to a 3-dimensional grid of numbers, called $GL_3$ ) when it is "twisted" by a specific pattern (a Dirichlet character).

Here is the breakdown of what the authors did, using simple analogies.

1. The Goal: Finding the "Sweet Spot" (Subconvexity)

Mathematicians have a "rule of thumb" for how big these L-functions can get. This is called the convexity bound. Think of it like a speed limit sign on a highway. It tells you the maximum speed a car might go, but it's a very conservative estimate.

However, mathematicians suspect the cars actually go much slower. The "speed limit" they are trying to prove is called the subconvexity bound.

The Problem: Previous attempts to lower this speed limit were like trying to slow down a car by gently tapping the brakes. They worked, but not enough.
The Goal: The authors wanted to slam on the brakes and prove the car is going significantly slower than anyone thought possible, specifically when you change two things at once: the "size" of the pattern (called $q$ ) and the "frequency" of the wave (called $t$ ). This is called a hybrid bound.

2. The Tool: A Magical Mirror (Spectral Reciprocity)

To prove this, the authors didn't just look at the car (the L-function) directly. Instead, they built a magic mirror.

In math, this is called Spectral Reciprocity.

The Analogy: Imagine you want to know how loud a drum is in a noisy room. Instead of listening to the drum directly, you look at the reflection of the sound in a mirror.
The Trick: The authors found a formula that connects the "sound" of a complex 3D drum ( $GL_3$ ) to the "sound" of a much simpler 4D drum ( $GL_4$ ).
Why it helps: The 3D drum is hard to analyze. The 4D drum is easier. By translating the problem into the language of the 4D drum, they could calculate the answer much more precisely.

3. The Obstacle: The "Noise" of the Crowd

When they looked at the reflection (the dual moment), they encountered a massive crowd of Dirichlet characters (mathematical patterns).

The Problem: Most of these patterns are quiet, but a few are very loud (exceptional characters). If you just average the noise, the loud ones ruin your calculation.
The Solution: The authors realized that these loud patterns aren't random; they form a specific "clique" or a coset (a group within a group).
The Innovation: They developed a new way to measure the noise within these specific cliques. They proved that even within these loud groups, the average noise is much lower than previously thought. This is like realizing that even in a rock concert, if you stand in the right corner, the music is actually quite quiet.

4. The Result: Breaking the Record

By combining the Magic Mirror (reciprocity) with the Quiet Corner (the new bound on the noise), they achieved something incredible:

Stronger Limits: They proved that the L-functions are much smaller than the old "speed limit" suggested.
Simultaneous Control: They did this while changing two variables at once (the size of the pattern and the frequency), which is much harder than changing just one.
The "Natural Limit": The authors note that their result is likely the best possible result you can get using this specific "first moment" method (looking at the average). It's like hitting the theoretical maximum speed of a car given the engine's design.

5. Why Does This Matter?

You might ask, "Who cares about the speed of a mathematical wave?"

The Riemann Hypothesis: These L-functions are cousins of the famous Riemann Zeta function. Understanding their behavior helps us get closer to solving the biggest unsolved problem in math: the Riemann Hypothesis.
Number Theory: These bounds help mathematicians understand how prime numbers are distributed and how they interact with other patterns.
Real World: While abstract, these techniques often lead to better algorithms in cryptography and coding theory, which secure our digital world.

Summary in a Sentence

The authors built a mathematical "magic mirror" to translate a difficult 3D problem into an easier 4D one, and then used a clever new way to filter out the "noise" in the data, allowing them to prove that these mathematical waves are much smaller and more predictable than anyone previously believed.

1. Problem Statement

The paper addresses the subconvexity problem for $L$ -functions associated with automorphic forms. Specifically, it seeks to establish bounds for the central values (and nearby values) of twisted $L$ -functions that are stronger than the "convexity bound" derived from the functional equation and Phragmén–Lindelöf principle.

The focus is on hybrid subconvexity, meaning the bounds must be uniform and strong simultaneously in multiple aspects:

The $q$ aspect: The conductor of the twisting Dirichlet character $\chi$ .
The $t$ aspect: The imaginary part of the argument $s = 1/2 + it$.
The spectral aspect: The eigenvalue of the automorphic form (either the Laplacian eigenvalue $t_f$ for Maass forms or the weight $k_f$ for holomorphic forms).

The specific objects of study are:

Degree 3: $L(s, F \otimes \chi)$ , where $F$ is a selfdual Hecke–Maaß cusp form for $SL_3(\mathbb{Z})$ and $\chi$ is a primitive Dirichlet character.
Degree 6: $L(s, F \otimes f)$ , where $f$ is a $GL_2$ cusp form (Maass or holomorphic) twisted by $\chi$ .

The authors aim to improve upon previous results (notably by Li, Blomer, and Petrow–Young) to reach what they term the "natural limit" of the first moment method, modulo current knowledge on second moments.

2. Methodology

The core of the proof relies on a novel spectral reciprocity formula that relates moments of $L$ -functions on the $GL_2$ side to moments on the $GL_1$ side.

A. The Spectral Reciprocity Formula ( $GL_3 \times GL_2 \rightleftharpoons GL_4 \times GL_1$ )

The authors prove an identity (Theorem 6.4) that equates a weighted sum (moment) of $GL_3 \times GL_2$ Rankin–Selberg $L$ -functions to a dual moment involving $GL_4 \times GL_1$ $L$ -functions.

Left-Hand Side (Primal): A moment of $L(1/2, F \otimes f \otimes \chi)$ over a family of $GL_2$ forms $f$ and an integral over the continuous spectrum.
Right-Hand Side (Dual): A moment of $L(1/2 + it, F \otimes \psi) L(1/2 - it, \psi)$ over Dirichlet characters $\psi$ , weighted by a specific character sum $Z_\chi(\psi; t)$ .
Derivation: The formula is derived by:
1. Starting with the Kuznetsov and Petersson trace formulas.
2. Applying the $GL_3$ Voronoï summation formula (Miller–Schmid/Blomer–Khan) to the Kloosterman terms.
3. Utilizing additive and analytic reciprocity to rearrange sums.
4. Performing analytic continuation from the region of absolute convergence ( $\Re(w) > 1$ ) to the central point $w = 1/2$ . This step is delicate and requires precise control over the test functions and the behavior of character sums.

B. The First Moment Method

The strategy follows the paradigm pioneered by Li and Blomer:

Non-negativity: The central $L$ -value $L(1/2, F \otimes f \otimes \chi)$ is non-negative because $F$ is selfdual. This allows the authors to bound a single $L$ -value by the average (moment) over a family.
Optimization: The spectral reciprocity formula introduces a "dual moment." To minimize the error term, the authors optimize the size of the family (via parameters $q_2$ and $U$ ) to balance the growth of the main terms against the dual moment.

C. Key Technical Inputs

Character Sums: A critical component is the analysis of the character sum $Z_\chi(\psi; t)$ , which arises from the reciprocity. The authors prove that for most characters, this sum exhibits square-root cancellation, while for exceptional characters (forming a coset), it is larger.
Hybrid Second Moment Bounds: To bound the dual moment, the authors invoke a Lindelöf-on-average upper bound for the second moment of Dirichlet $L$ -functions restricted to a coset. This relies on recent work by Petrow and Young.
Test Functions: The authors construct specific "admissible" test functions ( $h, h_{hol}$ ) that localize the spectral parameters to a specific range $[T-U, T+U]$ while ensuring rapid decay outside this range.

3. Key Contributions and Results

Main Theorem (Theorem 1.1)

The paper establishes strong hybrid subconvex bounds for $L(1/2 + it, F \otimes \chi)$ and $L(1/2, F \otimes f)$ .

Result: For a selfdual $F$ and primitive $\chi$ of conductor $q$ , with a divisor $q_1 | q$ such that $(q_1, q/q_1)=1$ :
$L\left(\frac{1}{2} + it, F \otimes \chi\right) \ll_{F, \epsilon} (q(|t|+1))^{\frac{3}{5} + \epsilon}$
(under optimal conditions on $q_1$ ).
Improvement: The convexity bound is $(q(|t|+1))^{3/4+\epsilon}$ . The authors improve the exponent from $3/4$ to $3/5$ in the $q$ -aspect (when $q$ has a suitable divisor) and $3/5$ in the $t$ -aspect.
Significance: This bound is "strong" in the sense that it represents the limit of what can be achieved using the first moment method, assuming the generalized Lindelöf hypothesis for the second moment of $GL_3$ $L$ -functions.

Theorem 1.2 (Moment Bounds)

The paper proves the underlying moment bound (Theorem 1.2) which serves as the engine for the subconvexity results. It provides an upper bound for the average of $L$ -values over a family, showing:
$\sum L(1/2, F \otimes f \otimes \chi) \ll (q_1 q_2 T U) (q_1 q_2 T)^\epsilon + (q_1 T)^{5/4} q_2^{1/2} U^{-1/4} (q_1 q_2 T)^\epsilon$
This bound is sharp enough to yield the $3/5$ exponent when optimized.

Eisenstein Case (Section 11)

The authors extend their method to the case where $F$ is replaced by a minimal parabolic Eisenstein series. This yields Weyl-strength subconvex bounds ( $1/6$ in the $t$ -aspect and $1/3$ in the $q$ -aspect) for $L$ -functions of the form $L(1/2, f \otimes \chi)$ , recovering and extending results by Petrow and Young.

4. Significance and Impact

Optimality of the Method: The paper demonstrates that the "first moment method" combined with spectral reciprocity can achieve exponents ( $3/5$ ) that were previously thought to be the barrier for this technique. It clarifies that further improvements would require breaking the current bounds on the second moment of $GL_3$ $L$ -functions.
Spectral Reciprocity: The explicit $GL_3 \times GL_2 \rightleftharpoons GL_4 \times GL_1$ reciprocity formula is a major structural contribution. It generalizes previous results (like Motohashi's formula for $GL_2$ ) to higher rank groups and provides a powerful tool for future investigations into $L$ -function moments.
Hybrid Uniformity: Unlike many previous results that treated the $q$ and $t$ aspects separately or with restrictions (e.g., $q$ being prime or square-free), this work provides bounds that are uniform and strong in both aspects simultaneously, even for non-square-free conductors.
Technical Innovation: The handling of the analytic continuation of the spectral reciprocity formula to the central point, particularly the management of the secondary main terms and the character sum $Z_\chi$ , represents a significant technical advancement in analytic number theory.

In summary, this paper pushes the boundaries of subconvexity for $GL_3$ $L$ -functions by introducing a sophisticated spectral reciprocity framework, achieving the strongest possible bounds currently accessible via moment methods, and providing a unified treatment of the conductor and spectral parameters.