Survey on Cremona groups from a median geometric point of view

Basic Information

• Paper ID: 2510.10838
• Title: Survey on Cremona groups from a median geometric point of view
• Author: Anne Lonjou (Laboratoire de mathématiques d'Orsay, Université Paris-Saclay)
• Classification: math.GR (Group Theory), math.AG (Algebraic Geometry)
• Publication Date: October 14, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.10838

Abstract

This is a survey article based on the author's lecture notes from the mini-course "Cremona groups and CAT(0) cube complexes" at the 2023 Riverside Geometric Group Theory Workshop. The article is aimed at both geometric group theorists and algebraic geometers, introducing recent constructions of actions of Cremona groups on median graphs.

Research Background and Motivation

Core Problem

The article centers on an important open problem:

Question 1.1: Let G be a finitely generated subgroup of Bir(X) such that every g ∈ G is regularizable. Is G regularizable? That is, does there exist a variety Y and a birational map φ: Y → X such that G is conjugate to a subgroup of the automorphism group of Y?

Research Motivation

1. Limitations of geometric group theory methods: Although geometric group theory methods have been highly effective in studying 2-dimensional Cremona groups, until recently these methods could not be applied to higher-dimensional Cremona groups due to the lack of non-trivial actions on geometric spaces.
2. Need for a unified framework: A unified geometric framework is needed to study birational transformation groups in different dimensions.
3. Theoretical extension: Extending classical geometric group theory results through concrete actions of large groups (non-finitely generated) on non-locally compact or infinite-dimensional geometric objects.

Core Contributions

1. Introduction of recent constructions of Cremona group actions on median graphs, including:
   • Blow-up graph Cb(S)
   • Rational blow-up graph Cb,k(S)
   • Jonquières graph
   • Graph Cℓ(X) for higher dimensions
2. Establishment of a dictionary between median geometry and birational geometry, connecting:
   • Elliptic isometries and regularizable elements
   • Distance and number of base points
   • Translation length and dynamical base point number
3. Resolution of regularization problems in specific cases:
   • 2-dimensional Cremona groups over finite fields
   • Jonquières groups
4. Construction of morphisms from certain Cremona groups to Z, providing new research tools for higher-dimensional cases

Detailed Methods

Foundational Theory of Median Graphs

Definition of Median Graphs

A graph X is a median graph if and only if every triple of vertices x₁, x₂, x₃ has a unique median point m satisfying:

d(xᵢ, xⱼ) = d(xᵢ, m) + d(m, xⱼ) for all i ≠ j

Relationship with CAT(0) Cube Complexes

Theorem 2.4: A graph is a median graph if and only if it is the 1-skeleton of some CAT(0) cube complex.

Blow-up Graph Construction

Definition of Marked Surfaces

A marked surface (T, φ) is a pair where T is a smooth projective surface and φ: T → S is a birational map. Two marked surfaces are equivalent if and only if the map between them is an isomorphism.

Definition of Blow-up Graph

Definition 4.1: The blow-up graph Cb(S) has vertices given by equivalence classes of marked surfaces (T, φ), with two vertices adjacent if and only if one is obtained from the other by blowing up a point.

Key Properties

Theorem 4.5: The blow-up graph is a median graph.

Proposition 4.11: Establishes correspondence between geometric and birational concepts:

1. Elliptic isometries ↔ projectively regularizable elements
2. Distance ↔ number of base points: d((S, id), (S, f)) = 2|B(f)|
3. Translation length ↔ dynamical base point number: ℓ(f) = 2μ(f)

Higher-Dimensional Generalization

Construction of Graph Cℓ(X)

For ℓ-marked varieties (Y, φ) where φ: Y → X is an isomorphism of codimension ℓ, construct graph Cℓ(X):

• Vertices: equivalence classes of ℓ-marked varieties
• Edges: connected by removing irreducible subvarieties of codimension ℓ+1

Theorem 5.6: The graph Cℓ(X) is a median graph.

Core Results

Study of Purely Elliptic Actions

Theorem 2.17 (GLU24): If G is a finitely generated group acting purely elliptically on a locally finite-dimensional median graph G, then G has bounded orbits.

The application of this result to blow-up graphs is limited because blow-up graphs are not locally finite-dimensional.

Regularization Results over Finite Fields

Theorem 4.18 (GLU24): Let F be a finite field and S a smooth projective surface defined over F. If G is a finitely generated subgroup of Bir(S) such that every element of G is projectively regularizable, then G is projectively regularizable.

Boundedness of Jonquières Groups

Theorem 4.22 (LPU24): A finitely generated subgroup of Bir(P²) consisting only of algebraic elements is bounded.

Morphisms from Higher-Dimensional Cremona Groups

Theorem 5.24 (GLU25): There exists a non-trivial homomorphism φ: Bir(Pⁿ) → ZDiv(Pⁿ)/≈ in the following cases:

1. n ≥ 5 and the field is infinite
2. n ≥ 4 and the field has characteristic 0
3. n = 3 and the field is a number field, a function field over a number field, or a function field over an algebraically closed field

Technical Innovations

1. Geometrization of Birational Concepts

Through the geometric structure of median graphs, abstract birational concepts (such as base points, exceptional loci, etc.) are geometrized, enabling the application of geometric group theory tools.

2. Unified Framework

Provides a unified framework for studying birational transformation groups in different dimensions, from surfaces to varieties of arbitrary dimension.

3. Use of Cubical Orientation

Through cubical orientation, group actions are connected with regularization problems, so that bounded orbits correspond to regularizability.

4. Multi-level Construction

By constructing different median graphs (Cb(S), Cb,k(S), Cℓ(X), etc.), appropriate geometric objects are provided for different problems.

Main Applications

Regularization Theory

1. Regularization of groups with FW property: Subgroups with FW property are projectively regularizable
2. Regularization of bounded groups: Bounded subgroups are projectively regularizable
3. Torsion and divisible elements: Groups generated by torsion and divisible elements are regularizable

Restrictions on Centralizers

Theorem 5.22: For f ∈ Bir(X) that is not codimension-1 pseudo-regularizable, its centralizer cent(f) either permutes fibers of some rational map, or contains ⟨f⟩ × H as a finite-index subgroup, where H is a torsion group.

Lower Bounds on Degree Growth

Theorem 5.23: For g ∈ Bir(Pᵈ) that is not codimension-1 pseudo-regularizable, the asymptotic growth of deg(gⁿ) is at least 1/(d+1)n.

Related Work

Geometric Group Theory Background

• Classical results of Serre on group actions on trees
• Development of CAT(0) cube complex theory
• Establishment of median graph theory

Cremona Group Research

• Pioneering work of Cantat-Lamy on hyperbolic actions of 2-dimensional Cremona groups
• Research of Blanc-Déserti on dynamical base point numbers
• Recent advances in higher-dimensional Cremona groups

Regularization Theory

• Work of Cantat-de Cornulier on commutator actions
• Research of Cornulier on partial actions

Conclusions and Discussion

Main Achievements

1. Successfully extended geometric group theory methods to the study of higher-dimensional Cremona groups
2. Established deep connections between birational geometry and median geometry
3. Resolved regularization problems in specific cases
4. Provided new research tools for higher-dimensional cases

Limitations

1. The general regularization problem (Question 1.1) remains open
2. Blow-up graphs are not locally finite-dimensional, limiting the application of certain results
3. Many problems in higher dimensions remain unsolved

Future Directions

1. Continue investigating regularization of 2-dimensional Cremona groups over general fields
2. Explore Tits alternatives for higher-dimensional Cremona groups
3. Study properties of more special subgroups
4. Develop new geometric tools for handling infinite-dimensional cases

In-Depth Evaluation

Strengths

1. Theoretical depth: Establishes deep connections between two important mathematical branches
2. Methodological innovation: First systematic application of median graph theory to birational geometry
3. Rich results: Obtains multiple important regularization results and structure theorems
4. Clear exposition: Detailed and accessible explanations for readers with different backgrounds

Weaknesses

1. Technical limitations: Some results apply only to special cases (e.g., finite fields, specific dimensions)
2. Open problems: The core problem (Question 1.1) remains unresolved in the general case
3. Computational complexity: Practical computation of median graphs and verification of properties may be difficult

Impact

1. Theoretical contribution: Provides a new research paradigm for birational geometry
2. Methodology: Demonstrates the powerful application of geometric group theory in algebraic geometry
3. Inspirational value: Provides new perspectives and tools for research in related fields

Applicable Scenarios

1. Theoretical research: Interdisciplinary research between birational geometry and geometric group theory
2. Concrete applications: Classification and property study of special Cremona groups
3. Teaching: Excellent material for connecting different mathematical branches

This survey provides a comprehensive and in-depth perspective on the geometric group theory properties of Cremona groups and is an important reference in this interdisciplinary field.