🔢 mathematics

De Rham logarithmic classes and Tate conjecture

This paper introduces De Rham logarithmic classes to establish a correspondence between logarithmic classes and algebraic cycles, ultimately proving the Tate conjecture for smooth projective varieties over fields of finite type over $\mathbb{Q}$ via an analytic result in the $p$ -adic setting.

Original authors: Johann Bouali

Published 2026-03-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Johann Bouali

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 hidden blueprint of a magnificent, complex building. In the world of mathematics, this "building" is a geometric shape called an algebraic variety. The "blueprint" is a set of mathematical rules that tell us which parts of the building are made of solid, physical bricks (algebraic cycles) and which parts are just shadows or illusions (abstract cohomology classes).

This paper, written by Johann Bouali, is a new guidebook for reading these blueprints. It introduces a special tool called "De Rham logarithmic classes" to solve a famous mystery known as the Tate Conjecture.

Here is the breakdown in simple terms:

1. The Two Languages of Geometry

To understand the building, mathematicians speak two different languages:

Language A (The Bricks): This describes actual, physical pieces of the building. If you have a wall, a floor, or a pillar, that's an "algebraic cycle." It's concrete.
Language B (The Shadows): This describes the "shape" or "vibe" of the building using calculus and topology. It's like measuring the wind patterns or the sound echoes in the building. This is called De Rham cohomology.

The Problem: Sometimes, the "Shadows" (Language B) look like they could be "Bricks" (Language A), but we can't prove it. The Tate Conjecture asks: If a shadow looks exactly like a brick, is it actually a brick?

2. The New Tool: "Logarithmic" Glasses

The author invents a pair of special glasses called De Rham logarithmic classes.

Think of the building's surface as having a special texture. Most mathematical tools look at the whole surface. But these "logarithmic" glasses only focus on the edges, corners, and singularities—the places where the building has sharp turns or where different parts meet.

The Discovery: The author proves that if you look at a "shadow" through these logarithmic glasses and it fits perfectly into the "brick" category (specifically, if it has a specific symmetry called "bidegree (d, d)"), then it is definitely a real brick.
The Analogy: Imagine you are trying to identify a real tree in a forest of plastic trees. Most trees look similar. But if you look at the roots (the logarithmic part), the real tree has a specific, messy, organic pattern, while the plastic one is smooth. The author proves that if the "roots" look right, the whole thing is a real tree.

3. The "No Ghosts" Rule

The paper also proves a "No Ghosts" rule.

If you look at a shadow that has a weird, mismatched shape (mathematically, a "bidegree (p, q)" where $p \neq q$ ), the logarithmic glasses show that it is empty. There is no brick there. It's just a ghost.
This simplifies the search: You only need to look for bricks in the "symmetric" zones.

4. Solving the Mystery (The Tate Conjecture)

The ultimate goal is to prove the Tate Conjecture for a specific type of building: one built over a field of numbers that looks like the rational numbers (fractions), but studied using "p-adic" numbers (a different way of counting that focuses on remainders).

The Strategy:

The Setup: The author takes a "shadow" (a Tate class) that is known to be invariant under the "Galois group" (a group of symmetries that shuffle the numbers around).
The Translation: Using a bridge called the p-adic comparison isomorphism, the author translates this "shadow" from the world of remainders (p-adic) into the world of calculus (De Rham).
The Logarithmic Check: Once translated, the author puts on the logarithmic glasses.
- The glasses reveal that this shadow is "logarithmic."
- Because of the main theorem proved earlier, being "logarithmic" and having the right shape means it must be a real algebraic cycle.
The Conclusion: The shadow was a brick all along! The Tate Conjecture is true for these buildings.

5. Why This Matters (The "Bad Reduction" Twist)

There is a tricky part. When you look at these buildings through the "p-adic" lens, the building might look like it has crumbled or changed shape (this is called "bad reduction").

The author shows that even if the building looks broken or has extra pieces when viewed through the p-adic lens, the "logarithmic" tool is smart enough to ignore the debris and identify the original, solid bricks underneath.
It's like looking at a shattered vase through a magnifying glass. The glass shows you thousands of shards. But the author's tool can tell you exactly which shards belong to the original vase and which are just dust, allowing you to reconstruct the vase perfectly.

Summary

Johann Bouali has created a new mathematical filter (logarithmic classes). He proved that this filter is so precise that it can distinguish between "real geometric objects" and "abstract mathematical illusions." By using this filter, he successfully solved a decades-old puzzle (the Tate Conjecture) for a wide range of geometric shapes, proving that if a shape behaves like a solid object in the right way, it is a solid object.

In one sentence: The author built a special pair of glasses that can see the "roots" of geometric shapes, proving that if a shape's roots look like a real building, then the whole thing is a real building, solving a major mystery in number theory.

1. Problem Statement

The paper addresses two fundamental problems in arithmetic geometry and the theory of motives:

Characterization of Algebraic Cycles: Can one characterize the De Rham cohomology classes of algebraic cycles purely in terms of "logarithmic" differential forms? Specifically, does the property of being a "logarithmic class" in De Rham cohomology imply that the class is algebraic?
The Tate Conjecture: Can the Tate conjecture (which posits that Galois-invariant $\ell$ -adic cohomology classes are algebraic) be proven for smooth projective varieties over fields of finite type over $\mathbb{Q}$ ? While the conjecture is known for $\ell \neq p$ , the case for $p$ -adic coefficients ( $\ell = p$ ) has historically been problematic due to the lack of a suitable comparison isomorphism that preserves the "algebraic" nature of classes in the $p$ -adic setting.

The author aims to bridge these by introducing De Rham logarithmic classes and showing that for varieties over fields of finite type over $\mathbb{Q}$ , the $p$ -adic Tate classes correspond to logarithmic classes, which are then proven to be algebraic.

2. Methodology

The paper employs a sophisticated blend of motivic homotopy theory, pro-étale cohomology, and $p$ -adic Hodge theory.

Logarithmic Forms: The author defines a subcomplex $\Omega^\bullet_{X, \log} \subset \Omega^\bullet_X$ consisting of forms generated by $df/f$ (logarithmic differentials). Unlike Deligne's logarithmic forms along a divisor, these are defined globally on the variety.
Motivic Properties: A crucial step is establishing that the presheaves of logarithmic forms admit transfers and are $\mathbb{A}^1$ -invariant. By invoking Voevodsky's Theorem, the author deduces that the cohomology of these presheaves is $\mathbb{A}^1$ -local. This allows the use of motivic machinery (purity isomorphisms, excision) to manipulate logarithmic classes.
Purity and Induction: The proof strategy relies on a finite induction on the codimension of the support of a class. The author proves that the purity isomorphism (relating cohomology with support to the cohomology of the normal bundle) preserves logarithmic classes. This reduces the problem to the case of divisors, where logarithmic classes correspond to the Picard group ( $H^1_{et}(X, \mathcal{O}^*_X)$ ).
$p$ -adic Comparison: For the Tate conjecture, the author utilizes the $p$ -adic crystalline comparison isomorphism (Fontaine-Messing/Bhatt-Scholze). They show that a Tate class (Galois invariant) maps to a logarithmic De Rham class via this isomorphism.
Integral Models and Desingularization: To handle the fact that integral models of cycles may have bad reduction (reducing to multiple components modulo $p$ ), the author constructs specific finite surjective morphisms and desingularizations. This allows them to "lift" the structure of the reduction modulo $p$ back to the generic fiber, ensuring the coefficients of the resulting cycle are rational (invariant under the Galois group).

3. Key Contributions and Results

A. De Rham Logarithmic Classes (Characteristic Zero)

Definition: Introduced the subgroup of De Rham cohomology $H^j_{DR}(X)$ consisting of logarithmic classes, defined as the image of $H^j_{et}(X, \Omega^\bullet_{X, \log})$ under the inclusion into De Rham cohomology.
Theorem 2 (Main Characterization):
- Forward: The De Rham class of any algebraic cycle of codimension $d$ is a logarithmic class of bidegree $(d, d)$ .
- Converse: Any logarithmic class of bidegree $(d, d)$ in $H^{2d}_{DR}(X)$ is the class of an algebraic cycle of codimension $d$ .
- Vanishing: For smooth projective varieties, there are no non-trivial logarithmic classes of bidegree $(p, q)$ where $p \neq q$ . This implies that odd-degree De Rham cohomology contains no non-trivial logarithmic classes.

B. Analytic Logarithmic Classes ( $p$ -adic Case)

Proposition 9: Extended the results to smooth projective varieties over $p$ $p$ -adic fields with good reduction.
- Logarithmic analytic De Rham classes of type $(d, d)$ correspond to formal algebraic cycles.
- Vanishing holds for $p \neq q$ in the analytic setting as well.

C. The Tate Conjecture

Theorem 3: Proved the Tate conjecture with $\mathbb{Q}_p$ coefficients for smooth projective varieties over fields of finite type over $\mathbb{Q}$ $Q$ .
- Mechanism: A Tate class $\alpha \in H^{2d}_{et}(X_{\bar{k}}, \mathbb{Q}_p(d))^G$ is mapped via the $p$ -adic comparison isomorphism to a logarithmic De Rham class $w(\alpha)$ .
- By Theorem 2/Proposition 9, $w(\alpha)$ is the class of a formal cycle.
- Using a geometric construction involving finite covers and desingularization, the author shows that the coefficients of this formal cycle must be invariant under the absolute Galois group $G$ , thereby proving the cycle is defined over the base field (up to rational equivalence).
Corollary 2: As a consequence, the Standard Conjectures hold for smooth projective varieties over $\mathbb{C}$ , and the Hodge Conjecture is equivalent to the statement that Hodge classes are logarithmic De Rham classes.

4. Significance

Resolution of the $p$ -adic Tate Conjecture: This is the most significant contribution. The Tate conjecture for $p$ -adic coefficients was previously considered false in general (due to counterexamples involving non-isogenous elliptic curves with good reduction). However, Bouali shows that for varieties defined over fields of finite type over $\mathbb{Q}$ , the conjecture holds. The paper clarifies that the failure in the general $p$ -adic case (e.g., over $\mathbb{Q}_p$ itself) is due to the lack of "global" Galois invariance required to descend the formal cycle to the base field.
New Characterization of Algebraicity: The paper provides a purely cohomological characterization of algebraic cycles using logarithmic forms, bypassing the need for explicit cycle constructions in the intermediate steps.
Unification of Hodge and Tate: By linking logarithmic classes to both Hodge classes (via period maps) and Tate classes (via $p$ -adic comparison), the work offers a unified perspective on the algebraicity of cohomology classes across different cohomology theories.
Technical Innovation: The use of Voevodsky's motivic framework to prove properties of logarithmic forms (specifically $\mathbb{A}^1$ -invariance and transfers) is a novel application that simplifies the proof of purity for these specific classes.

5. Conclusion

Johann Bouali's paper establishes a deep connection between logarithmic differential forms and algebraic cycles. By proving that logarithmic classes in De Rham cohomology are precisely the classes of algebraic cycles, the author successfully deduces the Tate conjecture for smooth projective varieties over fields of finite type over $\mathbb{Q}$ . This result resolves a long-standing open problem in the $p$ -adic setting for this specific class of fields, while also providing new insights into the Hodge conjecture and the structure of motivic cohomology.