🔢 mathematics

Weber modular curves and modular isogenies

This paper investigates modular curves and polynomials defined by Weber functions, analyzing their group actions and parametrizations to characterize isogeny graphs of supersingular elliptic curves, with applications to efficient cryptographic computations and explicit Galois representations.

Original authors: Leonardo Colò, David Kohel

Published 2026-04-01

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Leonardo Colò, David Kohel

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 a master locksmith trying to navigate a vast, invisible maze made of mathematical shapes called elliptic curves. These shapes are the backbone of modern cryptography (the code protecting your bank account) and deep number theory.

To move through this maze, you need keys. In math, these keys are called isogenies—special bridges that connect one curve to another. The paper you shared is like a new, ultra-efficient map and a set of super-lightweight keys that make navigating this maze much faster and easier.

Here is the breakdown of the paper's big ideas, translated into everyday language:

1. The Problem: The Maze is Too Heavy

Traditionally, mathematicians use a "master key" called the $j$ -invariant to describe these curves. Think of the $j$ -invariant as a giant, heavy, complex blueprint.

The Issue: When you try to calculate the bridges (isogenies) between curves using this heavy blueprint, the numbers get huge. It's like trying to carry a boulder up a mountain just to take a single step. The calculations are slow, and the data takes up a lot of space.

2. The Solution: Weber's "Lightweight" Keys

The authors introduce a new set of tools based on Weber functions.

The Analogy: If the old $j$ -invariant is a heavy boulder, the Weber functions are feather-light origami.
The Magic: These functions are "modular functions" (special mathematical recipes) that describe the same curves but with much smaller numbers.
- Size Reduction: The numbers involved are roughly 72 times smaller than the old method.
- Sparseness: The equations are "sparse," meaning they have fewer terms. It's like writing a sentence with only the essential words, removing all the fluff. This makes computer calculations lightning-fast.

3. The Map: Weber Curves and Fermat Curves

The paper describes specific shapes (curves) where these lightweight keys live.

The Connection: The authors show that these "Weber curves" are secretly twins of famous shapes called Fermat curves (think of the equation $x^n + y^n = z^n$ ).
Why it matters: Because they are twins, the authors can use the known, beautiful symmetries of Fermat curves to navigate the Weber curves. It's like realizing that a complex, winding forest path is actually just a straight line if you look at it from a different angle.

4. The Superpower: The "Supersingular" Shortcut

In cryptography, there is a special type of curve called a supersingular curve. These are the "VIP lanes" of the maze used in post-quantum cryptography (security against future quantum computers).

The Discovery: The authors prove that all the "supersingular" points on their Weber maps live in a very simple, predictable neighborhood (a field called $\mathbb{F}_{p^2}$ ).
The Benefit: You don't need to build a bigger, more complex map to find these VIP lanes. They are already right there, fully formed. This makes it incredibly efficient to generate the cryptographic keys needed for protocols like CSIDH (a new type of secure internet traffic).

5. The Application: Building Bridges and Finding Patterns

The paper does two main things with these tools:

Cryptographic Speed: They show how to build chains of bridges (isogenies) between curves much faster. This is crucial for making encryption that is both secure and fast enough for real-world use.
Mathematical Treasure Hunting: They use these maps to find hidden patterns in numbers (Galois representations). By using these lightweight keys, they can count and categorize thousands of mathematical objects that were previously too big or complex to study. It's like using a high-powered telescope to see stars that were previously too dim to spot.

Summary Analogy

Imagine you are trying to cross a river.

The Old Way: You try to swim across using a heavy lead vest (the old $j$ -invariant). It's exhausting, slow, and dangerous.
The New Way (This Paper): The authors give you a hoverboard (Weber functions).
- It's lighter (smaller numbers).
- It's faster (sparse equations).
- It has a GPS (the connection to Fermat curves) that tells you exactly where the safe, shallow spots (supersingular points) are.

The Bottom Line: This paper provides a new, highly efficient toolkit for mathematicians and cryptographers to navigate the complex world of elliptic curves, making secure communication faster and helping us discover new mathematical secrets that were previously out of reach.

1. Problem Statement

The paper addresses two primary challenges in the intersection of number theory and cryptography:

Efficient Isogeny Computation: Cryptographic protocols (such as SIDH and CSIDH) rely on computing isogenies between elliptic curves. This requires evaluating modular polynomials $\Phi_\ell(x, y)$ , which relate the $j$ -invariants of $\ell$ -isogenous curves. Standard modular polynomials (based on the $j$ -invariant) suffer from large coefficient sizes (height) and high density, making computations expensive.
Explicit Galois Representations: There is a need to construct explicit models for Galois representations associated with spaces of modular forms, particularly those involving supersingular elliptic curves and specific ramification at small primes (2 and 3).

The authors seek to utilize Weber functions (modular functions of level 48) to define modular curves that offer superior properties for these tasks compared to classical models.

2. Methodology

The authors employ a combination of algebraic geometry, modular form theory, and explicit computation:

Weber Functions and Level Structure: They focus on the classical triple of Weber functions ( $f, f_1, f_2$ ) of level 48. These functions generate a degree-72 cover of the $j$ -line ( $X(1)$ ). The authors analyze the action of $SL_2(\mathbb{Z})$ on these functions and their conjugates.
Construction of Modular Curves: They define explicit models for Weber curves ( $W_n$ ) and Fermat curves ( $F_n$ ) as quotients of the modular curve $X(48)$ . Specifically, they establish isomorphisms between certain Weber curves and Fermat curves (e.g., $W_n \cong F_n$ for $n|8$ ).
Derivation of Modular Polynomials: They derive Weber modular polynomials $\Phi_\ell(x, y)$ associated with these functions. A key methodological step is analyzing the symmetry properties of these polynomials under the action of roots of unity to prove sparsity (fewer non-zero terms) and height reduction (smaller coefficients).
Isogeny Chain Construction: The authors construct explicit families of elliptic curves parametrized by these modular curves. They detail the construction of 2-isogeny chains using the Fermat model, providing explicit equations for the curves and the isogeny maps.
Supersingular Analysis: They investigate the field of definition for supersingular points on these curves, proving that supersingular invariants on the relevant quotients split completely over $\mathbb{F}_{p^2}$ .
Graph Theory and Hecke Operators: They apply Mestre's method of graphs to analyze the supersingular isogeny graphs associated with these curves, counting "pseudo elliptic factors" (eigenspaces of Hecke operators) to estimate the distribution of modular elliptic curves with specific conductor properties.

3. Key Contributions

A. Sparsity and Height Reduction of Modular Polynomials

The most significant theoretical contribution is the demonstration that Weber modular polynomials possess superior computational properties compared to classical $j$ -invariant polynomials:

Sparsity: Due to the transformation $\Phi_\ell(\zeta_{24}x, \zeta_{24}^\ell y) = \zeta_{24}^{\ell+1}\Phi_\ell(x, y)$ , the number of non-zero monomials in $\Phi_\ell(x, y)$ is reduced by a factor of roughly 24. Specifically, a coefficient for $x^i y^j$ is non-zero only if $i + \ell j \equiv \ell + 1 \pmod{24}$ .
Height Reduction: The coefficients are asymptotically 72 times smaller in height than those of the classical $j$ -invariant polynomials. This is conjectured to be the maximal possible reduction.
Examples: The paper provides explicit low-degree examples (e.g., $\Phi_5, \Phi_7, \Phi_{11}$ ) showing these sparse structures.

B. Explicit Models and Parametrizations

Curve Definitions: The authors provide explicit projective equations for Weber curves $W_n$ and Fermat curves $F_n$ . For instance, $W_n$ is defined by $X_0^n + X_1^n + X_2^n = 0$ and $X_0 X_1 X_2 = \sqrt{2}^m X_3^3$ .
Elliptic Curve Families: They define explicit families of elliptic curves $E_0/X_n$ and $E_1/X_n$ parametrized by these curves, including their $j$ -invariants, discriminants, and torsion points.
2-Isogeny Chains: They construct explicit 2-isogeny chains over the field $K(Y_8)$ (where $Y_8$ is a quotient of the Fermat curve $F_8$ ), providing closed-form equations for the isogenies $\phi_0, \phi_1, \phi_2$ .

C. Supersingular Splitting Properties

The paper proves that the supersingular points on the Weber curves (specifically $W_{24}$ and its quotients) split completely over $\mathbb{F}_{p^2}$ .

This is crucial for cryptographic applications (like CSIDH) because it eliminates the need for base field extensions to define the isogeny graphs, similar to the behavior of $j$ -invariants but with the added benefit of the Weber structure.
Theorem 13 establishes that supersingular Weber invariants are defined over $\mathbb{F}_{p^2}$ for any odd prime $p$ .

D. Galois Representations and Data Enumeration

The authors use the supersingular isogeny graphs on these curves to compute data for Galois representations.
They enumerate "pseudo elliptic factors" for levels $N = mp$ where $m$ is a high power of 2 or 3 (e.g., $256p, 288p$ ).
Result: They show that these curves allow for the enumeration of isogeny classes of elliptic curves with conductors exceeding typical database bounds (e.g., $> 2 \times 10^6$ ), which are difficult to study using standard level-1 or low-level modular curves.

4. Results

Polynomial Efficiency: The Weber modular polynomials $\Phi_\ell$ are significantly smaller and sparser than classical counterparts, making them highly efficient for isogeny computations in cryptographic protocols.
Isomorphism: The paper establishes explicit isomorphisms between Weber curves $W_n$ and Fermat curves $F_n$ for divisors $n$ of 8, linking modular function theory with classical Fermat geometry.
Field of Definition: The supersingular moduli on $W_{24}$ are confirmed to be defined over $\mathbb{F}_{p^2}$ , validating their use in supersingular isogeny-based cryptography without extra field extensions.
Data Statistics: Table 1 in the paper provides counts of pseudo elliptic factors for various levels ( $X_1$ through $X_{12}$ and $Y_1$ through $Y_8$ ). The data suggests that the number of isogeny classes for composite levels (high powers of 2 and 3) is substantial, offering new datasets for studying the distribution of modular elliptic curves.

5. Significance

Cryptographic Impact: The work provides a theoretical and practical foundation for optimizing isogeny-based cryptography (like CSIDH). The reduced height and sparsity of Weber polynomials directly translate to faster key generation and isogeny evaluation. The fact that supersingular points split over $\mathbb{F}_{p^2}$ simplifies the implementation of these protocols.
Number Theoretic Insight: By connecting Weber functions to Fermat curves and explicit Galois representations, the paper bridges the gap between abstract modular form theory and explicit arithmetic geometry.
Computational Number Theory: The ability to enumerate isogeny classes for high composite levels (e.g., $288p$ ) extends the reach of computational number theory beyond the limits of current databases (like LMFDB), allowing researchers to study elliptic curves with specific ramification properties at 2 and 3 that were previously inaccessible.

In summary, Colò and Kohel demonstrate that Weber modular curves offer a "sweet spot" of high symmetry, low coefficient complexity, and favorable supersingular properties, making them superior to classical models for both cryptographic applications and the explicit study of Galois representations.