🔢 mathematics

Motivic six-functor formalism for log schemes

This paper establishes the motivic six-functor formalism for fs log schemes by proving key properties such as exact base change, the projection formula, and Poincaré duality, while also defining associated homology theories and the category of Chow motives.

Original authors: Doosung Park

Published 2026-07-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Doosung Park

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 understand the shape of a city, but you can only see the buildings when the lights are on. In mathematics, there is a branch called algebraic geometry that studies shapes defined by equations. Usually, these shapes are like perfect, smooth sculptures. But in the real world, things often have edges, corners, or boundaries where the rules get messy. For decades, mathematicians struggled to study what happens when these shapes break or touch a boundary, like a river hitting a dam. To fix this, they invented a tool called "log geometry." Think of log geometry as adding a special "shadow" or "label" to the edges of a shape. This label tells the mathematician exactly how the shape behaves right at the edge, turning a messy, broken boundary into something they can still calculate with.

Once you have these labeled shapes, the next big challenge is to build a universal toolkit to measure them. Mathematicians use something called "motivic homotopy theory," which is like a super-powerful camera that can take pictures of these shapes from every possible angle and zoom level, turning them into data that can be compared. The goal is to have a set of six magical rules (called the "six-functor formalism") that let you move these pictures around, flip them, stretch them, and combine them without losing any information. This paper is about finally getting those six rules to work perfectly for these new "labeled" shapes, even when they have tricky boundaries.

The Paper: A New Rulebook for Labeled Shapes

In this paper, the author, Doosung Park, builds a complete and rigorous rulebook for studying these "log schemes" (shapes with special edge labels) using the six-functor toolkit. Before this work, mathematicians had pieces of the puzzle, but they didn't have a complete, working system that allowed them to move data back and forth between different shapes while keeping the math consistent. Park proves that this system works, establishing what is known as the "motivic six-functor formalism" for these specific types of shapes.

Think of the six functors as six different ways to manipulate a shape: you can pull it back, push it forward, cut it open, close it up, or twist it. The paper proves that for log schemes, these moves follow a strict set of laws. Specifically, Park shows that you can swap the order of operations (like moving a shape and then cutting it, versus cutting it and then moving it) without changing the result. This is called the "exact base change" property. He also proves the "projection formula," which ensures that when you combine two shapes, the math behaves like a well-organized library where books are always in the right place. Most importantly, he proves "Poincaré duality." In simple terms, this is a rule that says if you know the shape of a surface, you automatically know the shape of its "inside" or its "dual" version, just like knowing the front of a coin tells you about the back.

The paper also introduces new ways to count and measure these shapes. It defines "Borel-Moore homology," which is a method for counting holes and features in shapes that have boundaries (like a disk with an edge). The author shows that for these log shapes, the usual rules of counting don't always apply in the same way they do for smooth, boundary-less shapes. For instance, if you stretch a shape with a boundary, the number of holes doesn't always stay the same, which is a natural phenomenon for things with edges. The paper also defines "Chow motives," which are like the "atoms" of these shapes. By breaking complex log schemes down into these atoms, mathematicians can study them more easily.

One of the most exciting findings is how these new rules handle a specific, simple log shape called the "standard log point." In the world of regular shapes, you can only have certain combinations of dimensions and twists. But in this new log world, the author proves that you can have any combination of dimensions and twists. It's as if the log labels unlock a new dimension of possibilities that were previously locked away. The paper demonstrates this by constructing a specific example of a "toroidal model of an elliptic curve" (a fancy donut shape with a log label) and showing how it breaks down into a sum of these new, flexible atoms.

The author is very careful to distinguish between what is proven and what is just a guess. The main results—the six functors, the base change, the duality, and the definitions of the new homology theories—are all rigorously proven using the mathematical tools developed in previous papers by the author and others. However, the paper also suggests a strategy for comparing these new log motives with older theories, but it leaves the final proof of that equivalence as an open question for future work. The paper does not claim to solve every problem in the field, but it provides the solid foundation and the complete rulebook needed for others to build the next generation of discoveries. It confirms that the "log" approach is not just a clever trick, but a robust framework that can handle the messy, boundary-filled reality of algebraic geometry with the same precision as the smooth, idealized world.

Technical Summary: Motivic Six-Function Formalism for Log Schemes

Problem Statement
Log geometry, developed by Deligne, Faltings, Fontaine, Illusie, and Kato, provides an algebraic framework for attaching boundaries to schemes, facilitating the study of limit behaviors in families (e.g., semistable degenerations). While logarithmic motivic homotopy theory has been extended to handle non- $\mathbb{A}^1$ -invariant cohomology theories, a comprehensive Grothendieck six-functor formalism for fs (fine and saturated) log schemes remained incomplete. Specifically, there was a need to establish exact base change, projection formulas, and Poincaré duality within the context of log schemes, particularly to define and compute Borel-Moore homology, G-theory, and Chow homology in a manner consistent with the logarithmic structure. Previous attempts, such as log Chow groups and prelog Chow rings, lacked certain invariance properties (e.g., $\square$ -invariance) or differed from the motivic framework established in the author's prior work.

Methodology
The paper operates within the framework of a "log motivic $\infty$ -category" $T$ , building upon the author's previous series of papers ([22], [23], [24], [25], [26]). The core methodology involves:

Extension of the $\infty$ -Category: The author utilizes the functor $T^{ex}: (lSch/B)^{op} \to CAlg(Pr^L)$ , which generates a full subcategory of the log motivic $\infty$ -category via colimits of motives of exact log smooth morphisms.
Purity Transformations: A central technical tool is the definition and analysis of purity transformations $p_f: f^\sharp \to f_! \Sigma_f$ for separated exact log smooth morphisms $f$ . The paper establishes conditions under which these transformations are isomorphisms (i.e., when $f$ is "pure").
Virtual Isomorphisms: The paper introduces and exploits the concept of "virtual isomorphisms" (morphisms that are isomorphisms of schemes and induce isomorphisms on the sheaves of groups of units $M^{gp}$ ). Theorem 3.2.6 proves that for such morphisms, the pullback functor $f^*$ is an equivalence of $\infty$ -categories. This is crucial for reducing global problems to local toric models.
Toric Stratification: The proof of Poincaré duality relies heavily on the stratification of affine toric log schemes. By analyzing the structure of monoids and their associated log schemes (specifically via charts and faces), the author reduces the general case to specific toric morphisms where duality can be verified inductively.
Compactification and Base Change: The formalism assumes the existence of Nagata compactifications and utilizes properties of compactifiable morphisms to define the exceptional direct image functor $f_!$ .

Key Contributions and Results

Establishment of the Six-Functor Formalism: The main theorem (Theorem 1.3.1) establishes the Grothendieck six-functor formalism for the functor $T^{ex}$ under the assumption that motives of vertical exact log smooth morphisms are compact. The formalism includes:
- Exact Base Change: For a cartesian square where one map is exact and the other is compactifiable, the natural transformation $g^* f_! \to f'_! g'^*$ is an isomorphism.
- Verdier Duality: For separated morphisms, $f_!$ admits a right adjoint $f^!$ .
- Poincaré Duality: For separated vertical exact log smooth morphisms $f$ , there is a natural isomorphism $f^\sharp F \simeq f_!(F \otimes Th(T_f))$ , where $Th(T_f)$ is the Thom motive of the tangent bundle.
- Lefschetz Duality: A specific duality is established for log smooth morphisms where the target has a trivial log structure, relating $f^!$ to the open immersion of the interior $j: X - \partial X \to X$ .
Definition of Homology and Cohomology Theories:
- Borel-Moore Homology: The paper defines $E^{BM}(X/S)$ for any morphism $f: X \to S$ of fs log schemes. Unlike standard motivic cohomology, Borel-Moore homology is not $\mathbb{A}^1$ -invariant nor invariant under verticalization, reflecting the behavior of topological Borel-Moore homology on manifolds with boundary.
- Specific Theories: The framework yields definitions for motivic cohomology, motivic Borel-Moore homology, Chow cohomology, Chow homology, homotopy K-theory, and G-theory for fs log schemes.
- Distinction of Chow Groups: The author emphasizes that Chow cohomology and Chow homology must be distinguished even for log smooth schemes if the boundary is non-empty, as Poincaré duality fails in the standard sense (replaced by Lefschetz duality).
Chow Motives over Log Schemes:
- The category of Chow motives $Chow(S)$ is defined as the idempotent completion of the homotopy category of motives generated by projective exact vertical log smooth morphisms.
- The Standard Log Point: A significant result (Proposition 1.4.1 and Proposition 4.3.6) demonstrates that the category $Chow(pt_{\mathbb{N}, k})$ (over the standard log point) is richer than the classical category $Chow(k)$. Specifically, $1(d)[n]$ exists in $Chow(pt_{\mathbb{N}, k})$ for all integers $d$ and $n$ , whereas in the classical case, $n$ must equal $2d$ .
- Example: The paper computes the motive of a toroidal model of an elliptic curve over $pt_{\mathbb{N}, k}$ , showing a decomposition $M(E) \simeq 1 \oplus 1[1] \oplus 1(1)[1] \oplus 1(1)[2]$ , which mirrors the Hodge diamond of an elliptic curve.

Significance and Claims
The paper claims to provide the necessary foundational machinery to perform motivic calculations on log schemes using the six-functor formalism. Its significance lies in:

Unification: It unifies the treatment of cohomology and Borel-Moore homology for log schemes under a single formalism, extending the work of Morel-Voevodsky to the logarithmic setting.
Computational Power: By establishing exact base change and duality, the paper provides a "recipe" for computing these theories, which was previously lacking.
Refinement of Existing Theories: The author explicitly compares their definitions with existing log Chow groups (Barrott) and prelog Chow rings (Böning et al.), noting that their theories satisfy $\square$ -invariance and other properties that previous definitions lacked, thereby distinguishing them.
New Objects: The introduction of Chow motives over the standard log point reveals a new class of motivic objects ( $1(d)[n]$ for arbitrary $n$ ) that are not present in the classical theory, suggesting a deeper structure in the motivic world of log schemes.

The paper does not claim to solve the Cst conjecture or construct specific new cohomology theories from scratch but rather provides the formal infrastructure (the six-functor formalism) that allows such theories to be defined, manipulated, and compared rigorously within the context of fs log schemes.

More like this