🔢 mathematics

Asymptotic Betti bounds for hypersurfaces in a singular variety

This paper establishes asymptotic upper bounds on the total Betti numbers of degree $d$ hypersurfaces within a possibly singular projective variety $X$ , showing that these bounds are proportional to $\deg(X) \cdot d^n$ with improved sharpness when $X$ is a local complete intersection, and extends these results to general constructible sheaves.

Original authors: Xuanyu Pan, Dingxin Zhang, Xiping Zhang

Published 2026-01-29

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Xuanyu Pan, Dingxin Zhang, Xiping Zhang

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 an architect trying to understand the "shape" of a building. In mathematics, specifically in a field called algebraic geometry, these "buildings" are called varieties. They are shapes defined by equations. Sometimes these shapes are smooth and perfect, like a polished marble sphere. Other times, they are crumpled, jagged, or have sharp corners and holes—they are singular.

The paper you provided is like a new set of blueprints that helps mathematicians count the "rooms" and "corridors" inside these shapes, even when they are built on a very rough, uneven foundation.

Here is a breakdown of what the authors, Xuanyu Pan, Dingxin Zhang, and Xiping Zhang, have achieved, using simple analogies.

1. The Goal: Counting the "Holes" in a Shape

In topology (the study of shapes), we care about things like holes, loops, and voids. Mathematicians call the total count of these features the Betti number.

The Analogy: Imagine a donut. It has one hole. A sphere has zero holes. A pretzel has three. The Betti number is just a scorecard for how complex the shape is.
The Problem: If you take a smooth, perfect shape (like a sphere) and slice it with a knife (a "hypersurface"), you can predict how many new holes appear based on how sharp or deep the cut is (the "degree" of the cut).
The Challenge: What happens if the original shape is a crumpled piece of paper or a jagged rock (a singular variety)? Standard rules break down. The authors wanted to find a rule that works even for the messiest, most broken shapes.

2. The Main Discovery: A "Safety Net" for Messy Shapes

The authors proved that no matter how ugly or singular the original shape ( $X$ ) is, if you slice it with a degree $d$ cut, the total number of holes in the resulting slice ( $Y$ ) cannot grow infinitely fast.

They found a mathematical speed limit (an upper bound).

The Formula: They showed the number of holes is roughly proportional to $d^n$ (where $n$ is the dimension of the shape).
The "Safety Net": They proved that even for the worst-case, most singular shapes, the number of holes is bounded by a specific formula:
$\text{Holes} \leq 3 \times (\text{Size of Original Shape}) \times d^n + \text{a small correction}$
Think of this as a safety net. Even if the shape is falling apart, the number of new holes created by the cut will never exceed this specific ceiling.

3. The "Local Complete Intersection" Bonus

The paper makes a special distinction between two types of messy shapes:

General Messy Shapes: These are just crumpled and jagged. For these, the "safety net" has a factor of 3 in front of it. It's a bit loose, but it holds.
Local Complete Intersections: These are shapes that are messy, but their messiness is "organized." Imagine a pile of sand that looks messy from afar, but if you zoom in, it's just a stack of neat layers.
- For these "organized messes," the authors found a tighter, more precise safety net. The factor of 3 disappears, and the bound becomes much sharper:
  $\text{Holes} \leq 1 \times (\text{Size of Original Shape}) \times d^n + \text{correction}$
- The Metaphor: It's like realizing that while a crumpled ball of paper has unpredictable tears, a stack of crumpled newspapers still follows a predictable pattern of tearing.

4. The Secret Weapon: "Perverse Sheaves"

How did they do it? They didn't just look at the shape; they looked at the "ghosts" inside the shape.

The Concept: In advanced math, there are objects called perverse sheaves. These aren't actual ghosts, but they act like "smart sensors" that can detect the structure of a shape even when it's broken.
The Trick: The authors realized that if they treated their counting problem as a problem about these "smart sensors" (perverse sheaves) rather than just the raw shape, they could use a powerful tool called the Weak Lefschetz Theorem.
The Analogy: Imagine trying to count the rooms in a collapsed building. If you just look at the rubble, it's impossible. But if you have a "smart sensor" that can see through the rubble and tell you where the walls used to be, you can count the rooms. The authors used these sensors to navigate the jagged parts of the shape.

5. The "Proper Degeneration" Lemma

This is the paper's most technical invention, but here is the simple version:

The Idea: Imagine you have a smooth shape and you slowly let it crumble into a jagged one. The authors proved that you can track the "holes" as the shape crumbles.
The Metaphor: Think of a clay sculpture. As you slowly squish it, the holes might merge or disappear. The authors proved that the number of holes in the final, squished version is always "controlled" by the holes in the smooth version plus a few extra terms. This allows them to take a known result for smooth shapes and apply it to jagged ones.

6. The "Universal" Constant (Independence of $\ell$ )

One of the most impressive parts of the paper is that their rules work regardless of the "lens" you use to look at the shape.

The Context: Mathematicians often use different number systems (like different prime numbers) to study shapes. Sometimes, a rule works for one number system but fails for another.
The Result: The authors proved that their "safety net" constants (the numbers in their formula) are universal. They don't change depending on which number system you use.
The Analogy: It's like finding a rule for how many bricks fall out of a wall when you hit it. The authors proved that this rule is the same whether you hit it with a hammer made of wood, steel, or plastic. The physics of the wall doesn't care what tool you use.

Summary

In short, this paper provides a universal rulebook for predicting the complexity of slices taken from any geometric shape, no matter how broken or jagged that shape is.

They established a maximum limit on how many "holes" can appear.
They found that for "organized" jagged shapes, this limit is tighter and more precise.
They used smart sensors (perverse sheaves) and a crumbling technique (degeneration) to prove it.
They proved these rules are universal, working for all mathematical "lenses."

This is a foundational result. It doesn't tell you how to build a bridge or cure a disease directly, but it gives mathematicians the confidence to count and measure the most chaotic shapes in the universe of algebraic geometry, knowing they will never exceed a predictable limit.

Technical Summary: Asymptotic Betti Bounds for Hypersurfaces in a Singular Variety

Problem Statement
The paper addresses the problem of establishing uniform upper bounds for the total Betti numbers of hypersurfaces within a projective variety $X$ over an algebraically closed field $k$ . Specifically, for a projective variety $X \subseteq \mathbb{P}^N$ of dimension $n$ and a hypersurface $Y \subseteq X$ of degree $d$ , the authors seek to bound the quantity:
$B(Y, \mathbb{F}_\ell) = \sum_i \dim H^i(Y; \mathbb{F}_\ell)$
as $d \to \infty$ . The study focuses on two critical aspects: the asymptotic behavior of these bounds in terms of $d$ and the dependence of the constants on the auxiliary prime $\ell$ . While optimal bounds were previously established for hypersurfaces in smooth projective space $\mathbb{P}^N$ (e.g., by Milnor, Thom, Katz, and more recently by Hu and Teyssier), the behavior for hypersurfaces in arbitrarily singular varieties remained an open question.

Methodology
The authors employ a strategy that shifts the focus from constant sheaves to general perverse sheaves, utilizing the theory of characteristic classes and degeneration techniques.

Shift to Perverse Sheaves: Rather than bounding the Betti numbers of the constant sheaf directly, the authors first establish bounds for the restriction of an arbitrary perverse sheaf $\mathcal{P}$ to a hypersurface. This allows for the application of the perverse weak Lefschetz theorem, which controls cohomology in most degrees. Bounds for constant sheaves are subsequently deduced via dévissage (decomposition into perverse cohomology sheaves).
Proper Degeneration Lemma: A central technical innovation is the Proper Degeneration Lemma (Lemma 4.1). Unlike the standard "perverse degeneration lemma" which controls middle-dimensional cohomology for affine morphisms, this new lemma relates the cohomology of a special (degenerating) hypersurface to both a general fiber and the ambient variety. It provides inequalities for cohomology in all degrees, not just the middle dimension, which is essential for handling singular varieties where the shifted constant sheaf is not necessarily perverse.
Characteristic Classes: The authors utilize Takeshi Saito's theory of characteristic classes for constructible complexes in the étale setting. They employ the Index Formula to compute Euler characteristics of hypersurface restrictions via characteristic cycles. This allows them to express the leading term of the Betti bound explicitly in terms of the degree of the support of the sheaf and the degree of the ambient variety.
Uniformity in $\ell$ : To address the dependence on the prime $\ell$ , the authors utilize the framework of compatible systems of sheaves. By working over finite fields and invoking the Grothendieck trace formula and properties of characteristic cycles under decomposition homomorphisms, they demonstrate that the Euler characteristics and characteristic numbers involved in the bounds are independent of $\ell$ .

Key Contributions and Results

The paper establishes the following main theorems:

General Singular Varieties (Theorem 1): For any $n$ -dimensional projective variety $X$ (possibly singular) and any degree $d$ hypersurface $Y \subseteq X$ , there exists a constant $C > 0$ independent of $d$ and $Y$ such that:
$B(Y, \mathbb{F}_\ell) \leq 3 \deg(X) \cdot d^n + C \cdot d^{n-1}$
The factor of 3 arises from the technical difficulty of controlling cohomology in degrees where the perverse weak Lefschetz theorem does not apply directly when the sheaf is not perverse.
Local Complete Intersections (Theorem 2): If $X$ is an irreducible projective variety with at worst local complete intersection (LCI) singularities, the bound improves to:
$B(Y, \mathbb{F}_\ell) \leq \deg(X) \cdot d^n + C \cdot d^{n-1}$
In this case, the shifted constant sheaf is perverse, allowing for a sharper asymptotic estimate where the bound is asymptotically sharp ( $B(Y, \mathbb{F}_\ell) \sim \deg(X) d^n$ ).
Perverse Sheaf Restrictions (Theorem 3): The authors prove a more general result for any perverse sheaf $\mathcal{P}$ on a smooth projective variety $P$ with support of dimension $n$ . The total Betti number of its restriction to a degree $d$ hypersurface is bounded by $A \cdot d^n + B \cdot d^{n-1}$ , where $A$ and $B$ are explicit constants derived from the characteristic classes of $\mathcal{P}$ .
Uniformity in $\ell$ (Theorem 4): The authors prove that the constants $C$ in Theorems 1 and 2, and $A, B$ in Theorem 3, can be chosen to be independent of $\ell$ . This holds provided the sheaves are part of a compatible system (e.g., defined over a finite field).

Significance and Claims
The paper claims to provide the first uniform asymptotic upper bounds for the Betti numbers of hypersurfaces in arbitrarily singular projective varieties. The results generalize classical bounds known for $\mathbb{P}^N$ to the singular setting.

Sharpness: The authors note that for local complete intersections, the leading term $\deg(X) \cdot d^n$ is asymptotically sharp, matching the behavior in the smooth case. For general singular varieties, the bound is slightly weaker (due to the factor of 3 and the nature of the singularities) but remains "remarkably sharp" in its asymptotic order $d^n$ .
Methodological Advance: The work introduces the Proper Degeneration Lemma as a tool to handle cohomology in non-middle degrees for singular varieties, overcoming limitations of previous methods that relied heavily on the weak Lefschetz theorem or Euler characteristic tricks which fail when perversity conditions are not met.
Arithmetic Application: The proof of $\ell$ -independence is significant for arithmetic algebraic geometry, particularly for applications in equidistribution theorems where uniformity across different primes is required.

The paper concludes by noting an open problem: whether the constant $C$ in the bound for general constructible complexes (not just sheaves) can be made independent of $\ell$ , a question that hinges on the compatibility of perverse cohomology functors with compatible systems.