🔢 mathematics

Cell decomposition and dual boundary complexes of character varieties

This paper establishes the weak geometric P=W conjecture for very generic $GL_n(\mathbb{C})$ -character varieties by employing an enhanced cell decomposition theorem and motivic cohomology techniques to inductively prove that their dual boundary complexes are homotopy equivalent to spheres of dimension $d-1$ .

Original authors: Tao Su

Published 2026-06-04

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Tao Su

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ 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 understand the shape of a very complex, invisible object. In mathematics, this object is called a Character Variety. Think of it as a giant, multi-dimensional map that records all the possible ways a specific set of rules (called "monodromies") can be arranged on a surface with holes (like a donut with punctures).

This paper, written by Tao Su, is about figuring out the "skeleton" or the "boundary shape" of these complex maps. Specifically, it tackles a famous guess (conjecture) made by mathematicians Katzarkov, Noll, Pandit, and Simpson, known as the Weak Geometric P=W Conjecture.

Here is the simple breakdown of what the paper does, using everyday analogies:

1. The Big Question: What does the edge look like?

Imagine you have a complex 3D sculpture made of glass. You can't see the inside clearly, but you can look at the boundary. The conjecture asks: If you take the "dual boundary complex" (a mathematical way of describing the skeleton of the edge) of these Character Varieties, does it look like a perfect sphere?

The Analogy: Think of the Character Variety as a giant, inflated balloon with many layers. The "dual boundary complex" is like the wireframe you would get if you popped the balloon and flattened the skin into a single, connected shape.
The Claim: The authors prove that for a very specific, "very generic" type of these balloons, this flattened wireframe is indeed homotopy equivalent to a sphere. In simple terms: The edge of this complex shape is topologically a sphere.

2. The Strategy: Breaking it down into Lego bricks

To prove this, the authors couldn't just look at the whole shape at once; it was too messy. They needed to take it apart.

The Cell Decomposition: The paper introduces a powerful new way to break the Character Variety into smaller, manageable pieces called "cells."
The Analogy: Imagine trying to understand a complex city. Instead of looking at the whole city at once, you break it down into neighborhoods.
- Some neighborhoods are just open squares (like a grid of streets).
- Some are like open fields.
- The authors proved that their "city" (the Character Variety) is built entirely out of these simple blocks.
- Crucially, they found that most of these blocks are "contractible," meaning if you squished them, they would turn into a single point. In topology, a shape that can be squished into a point doesn't add any "holes" or complex structure to the overall skeleton.

3. The "Very Generic" Condition

The paper focuses on a special case called "very generic."

The Analogy: Imagine you are arranging a deck of cards. If you arrange them randomly, you might get a messy pile. But if you arrange them so that no two cards share the same suit or number in a specific way (a "regular semisimple" arrangement), the pattern becomes much clearer and easier to predict.
The authors assume their mathematical "cards" (conjugacy classes) are arranged in this clean, non-repetitive way. Under this condition, the "city" of the Character Variety has a very clear structure.

4. The "Remove and Reduce" Trick

Once they broke the shape into Lego bricks, they used a clever trick to figure out the shape of the whole.

The Logic: They showed that if you remove a piece of the shape that is "simple" (like a contractible block), the overall "skeleton" of the remaining shape doesn't change.
The Result: By peeling away all the simple, contractible layers one by one, they were left with just one single, central piece: a giant open torus (which is like a donut shape).
The Conclusion: The "skeleton" of a giant donut is a sphere. Therefore, the skeleton of the original complex shape is also a sphere.

5. What about the "Zariski Cancellation Problem"?

In a side note, the paper touches on a famous unsolved puzzle in math called the Zariski Cancellation Problem.

The Analogy: Imagine you have a mysterious box. You know that if you attach a long, thin tube to it, the whole thing looks exactly like a standard cylinder. Does that mean the mystery box was just a cylinder to begin with?
The authors suggest that the shapes they found in their decomposition might be the first examples of "mystery boxes" that look like cylinders when you add a tube, but are actually different inside. This is a potential counterexample to the cancellation problem in higher dimensions.

Summary

Tao Su's paper proves that for a specific, well-behaved type of complex mathematical shape (Character Varieties), the "skeleton" of its boundary is a sphere. They did this by:

Dissecting the shape into simple, Lego-like blocks.
Showing that most of these blocks are topologically boring (they squish to a point).
Peeling them away to reveal that the core structure is a sphere.

This confirms a major prediction in modern geometry, linking the algebraic rules of the shape to its topological "skeleton."

Technical Summary of "Cell Decomposition and Dual Boundary Complexes of Character Varieties" by Tao Su

Problem Statement
The paper addresses the weak geometric P=W conjecture, proposed by Katzarkov, Noll, Pandit, and Simpson. This conjecture asserts that for a smooth Betti moduli space $M_B$ (a character variety) of complex dimension $d$ over a punctured Riemann surface, the dual boundary complex $D_{\partial}M_B$ is homotopy equivalent to a $(d-1)$ -dimensional sphere ( $S^{d-1}$ ).

While the full geometric P=W conjecture relates the "Hitchin fibration at infinity" on the Dolbeault side to a fibration intrinsic to the Betti side, the weak form focuses specifically on the topology of the dual boundary complex. Previous results established this for specific cases (e.g., $G=SL_2(\mathbb{C})$ , Painlevé cases, and certain singular or wild character varieties), but a general proof for higher-rank groups with generic monodromies remained open.

Methodology
The author employs a strategy combining cell decomposition techniques with motivic cohomology and diagram calculus. The core approach involves:

Refined Cell Decomposition: The paper improves upon A. Mellit's cell decomposition theorem for character varieties. Instead of decomposing a vector bundle over the variety, the author constructs a direct cell decomposition of the very generic character variety $M_\rho$ itself. This is achieved by:
- Introducing a diagram calculus of matrices (braid matrix diagrams) to relate character varieties to braid varieties.
- Utilizing the Bruhat decomposition and specific coordinate changes to reduce the defining equations of the character variety to those of restricted twisted braid varieties.
- Establishing that $M_\rho$ decomposes into locally closed affine subvarieties of the form $(\mathbb{C}^\times)^a \times A$ , where $A$ is a variety stably isomorphic to an affine space $\mathbb{A}^b$ .
Inductive Computation of Dual Boundary Complexes:
- The author utilizes a remove/reduction lemma (Lemma 3.7/3.8). This lemma states that if a smooth variety $X$ contains a closed subvariety $Z$ whose dual boundary complex is contractible, then the dual boundary complex of $X$ is homotopy equivalent to that of the complement $X \setminus Z$ .
- By ordering the cells in the decomposition, the author shows that all cells except the maximal dimensional one have dual boundary complexes that are contractible (due to the presence of affine space factors).
- The maximal cell is shown to be an open dense algebraic torus $(\mathbb{C}^\times)^d$ , whose dual boundary complex is known to be $S^{d-1}$ .
Motivic Characterization: The proof relies on a key property derived from the author's subsequent work [59]: if a variety $A$ is stably isomorphic to $\mathbb{A}^b$ (for $b \ge 1$ ), its dual boundary complex is contractible. This is established via a motivic characterization of the integral cohomology of dual boundary complexes.

Key Contributions and Results

Theorem 0.5 (Theorem 2.10): The paper establishes a strong form of the cell decomposition for "very generic" $GL_n(\mathbb{C})$ -character varieties. A variety is "very generic" if the local monodromies are specified by generic semisimple conjugacy classes, with at least one being regular semisimple.
- The decomposition is $M_\rho = \bigsqcup M_\rho(\vec{w}, p)$ , where each stratum is isomorphic to $(\mathbb{C}^\times)^a \times A_\rho(\vec{w}, p)$ .
- Crucially, $A_\rho(\vec{w}, p)$ is stably isomorphic to $\mathbb{A}^b$ .
- There exists a unique maximal stratum (corresponding to a specific walk in the braid group) which is an open dense torus $(\mathbb{C}^\times)^d$ .
Theorem 0.4 (Theorem 3.10): The main result proves the weak geometric P=W conjecture for all very generic character varieties.
- For a very generic $M_\rho$ of dimension $d$ , the dual boundary complex $D_{\partial}M_\rho$ is homotopy equivalent to $S^{d-1}$ .
- The proof proceeds by inductively removing lower-dimensional strata (which have contractible dual boundary complexes) until only the maximal torus remains.
Connection to the Zariski Cancellation Problem: The paper notes that the varieties $A_\rho(\vec{w}, p)$ appearing in the decomposition are stably isomorphic to affine spaces but are not necessarily isomorphic to them. The author suggests these provide potential systematic counterexamples to the Zariski cancellation problem in characteristic zero for dimensions $\ge 3$ , though this is presented as an observation rather than a resolved theorem.

Significance and Claims
The paper claims to resolve the weak geometric P=W conjecture for the class of very generic character varieties. This extends previous results which were limited to rank 2 or specific low-dimensional cases.

The significance lies in:

Topological Verification: It confirms that the dual boundary complex of these moduli spaces has the homotopy type of a sphere, supporting the broader geometric P=W philosophy that the topology of the Betti moduli space at infinity mirrors the topology of the Hitchin base.
Methodological Advancement: It provides a self-contained, rigorous cell decomposition for the character variety itself (not just a bundle over it), utilizing a novel diagram calculus that connects character varieties to braid varieties more directly than previous approaches.
Motivic Insight: The work bridges the gap between the rational cohomology (often studied via the P=W conjecture) and integral cohomology of the dual boundary complex, utilizing motivic weight cohomology to establish contractibility.

Limitations and Open Problems
The author explicitly states that the result holds for very generic varieties (where at least one conjugacy class is regular semisimple). The case where the variety is only generic (but not very generic) remains open.

For the generic case, the paper notes that $D_{\partial}M_B$ is known to be a rational homology sphere, but proving it is a topological sphere requires showing that the integral cohomology is torsion-free and that the fundamental group is trivial.
The paper identifies two specific obstructions for the general generic case: (I) proving the integral cohomology of the dual boundary complex is torsion-free, and (II) proving the fundamental group is trivial.
The author speculates that a motivic improvement of the degeneration arguments used in the Curious Hard Lefschetz theorem might be necessary to resolve these integral cohomology questions.