🔢 mathematics

Derived algebras on formal stacks and prismatic gauges

This paper investigates the interaction between derived algebras and formal derived geometry, specifically within the context of prismatization, to establish classification theorems for derived algebras on filtered formal stacks and related classifying stacks.

Original authors: Shubhankar Sahai

Published 2026-03-02

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Shubhankar Sahai

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

The Big Picture: Building a Universal Toolkit for "Fuzzy" Shapes

Imagine you are an architect trying to design a building, but the materials you have are a bit strange. They aren't just solid bricks; they are "fuzzy," "ghostly," or "shifting" materials that change depending on how you look at them. In mathematics, these materials are called Derived Algebras. They are used to describe shapes and spaces that have hidden layers of complexity, like a hologram that reveals more details the closer you get.

This paper is about building a new, robust toolbox to work with these fuzzy shapes, specifically when they are "formal" (meaning they are infinitely zoomed-in versions of a shape, like looking at a pixel so closely it becomes a whole world) and when they involve "prisms" (a special kind of mathematical lens used to study numbers).

The author, Shubhankar Sahai, is essentially saying: "We have these powerful tools for fuzzy shapes, but we didn't know how to use them when the shapes are infinitely zoomed-in or filtered. This paper teaches us how to use the tools in those specific, tricky situations."

Key Concepts Explained with Analogies

1. Derived Algebras: The "Ghostly" Building Blocks

The Concept: In normal math, a ring is like a set of numbers with rules for adding and multiplying. A Derived Algebra is like a "super-ring" that remembers not just the numbers, but also the history of how those numbers were made. It's like a 3D sculpture where the shadows cast by the object tell you as much about the object as the object itself.
The Analogy: Imagine a normal ring is a photograph of a tree. A derived algebra is a VR simulation of that tree. If you walk around it, you see the leaves, the bark, and the roots. If you "zoom in" (mathematically), you see the cellular structure. It captures the "fuzziness" or uncertainty of the shape that a flat photo misses.

2. Formal Stacks: The "Infinite Zoom"

The Concept: A Formal Stack is a geometric object that is "completed" along a specific direction. Think of it as taking a shape and zooming in on a specific point forever.
The Analogy: Imagine you are looking at a map of a city. A Formal Stack is like taking a magnifying glass and zooming in on a single street corner until the entire universe is just that street corner. You aren't looking at the whole city anymore; you are looking at the "infinitesimal neighborhood" of that corner. The paper studies how our "ghostly" building blocks (Derived Algebras) behave when they live inside these infinite zooms.

3. Prismatic Gauges: The "Mathematical Prism"

The Concept: This comes from Prismatic Cohomology, a hot topic in modern math that tries to unify different ways of studying numbers (like how water can be ice, liquid, or steam). A "Prism" is a pair of a ring and an ideal that acts like a lens.
The Analogy: Imagine white light (a complex number system) hitting a glass prism. The prism splits the light into a rainbow (different components). A Prismatic Gauge is a way of organizing these split colors so you can study them together. The paper shows how to build these "rainbow organizers" using our fuzzy building blocks.

4. The "Rees Construction": The "Time-Lapse Camera"

The Concept: This is a mathematical trick to turn a "filtered" object (one with layers, like an onion) into a "graded" object (one with distinct steps, like a staircase).
The Analogy: Imagine you have a stack of pancakes (a filtered object). The Rees Construction is like a time-lapse camera that takes a photo of the stack, then adds a "time" variable to the photo. Suddenly, the stack isn't just a pile; it's a 3D sculpture where the height represents the "time" or "layer." This paper proves that you can do this time-lapse trick even with your fuzzy, ghostly building blocks.

What Did the Author Actually Do?

The paper is very technical, but the main achievements can be summarized as three "New Rules" for the toolbox:

1. The "Completion" Rule (Section 2)

The Problem: When you zoom in infinitely (formal completion), your fuzzy building blocks sometimes break or disappear.
The Solution: The author proved that you can "fix" these broken blocks by completing them. He showed that the category of "completed fuzzy blocks" is a perfect, self-contained universe.
The Metaphor: It's like realizing that if you try to build a house out of sand, it collapses. But if you pour water on the sand (completion), it becomes solid concrete. The paper gives the recipe for turning "sand" into "concrete" for these specific shapes.

2. The "Symmetry" Rule (Section 3 & 4)

The Problem: These shapes often have symmetries (like a wheel spinning). The author wanted to know: "If I have a fuzzy block that spins, can I describe it just by looking at its stationary parts?"
The Solution: Yes! He proved that a "spinning fuzzy block" is exactly the same thing as a "stationary fuzzy block with a special label" (a coaction).
The Metaphor: Imagine a spinning top. It's hard to describe the motion. But if you take a photo of it and label the photo with "Spinning," you can describe the whole motion just by looking at the still photo and the label. The paper proves this works for the most complex, ghostly shapes.

3. The "Classification" Rule (Section 4)

The Problem: There are many different types of these formal stacks. How do we know which fuzzy block belongs to which stack?
The Solution: The author created a "dictionary." He showed that for a specific class of stacks (those related to the "Prism" and "Filtering"), there is a one-to-one match between the stacks and a specific type of filtered algebra.
The Metaphor: It's like having a library where every book (stack) has a unique barcode (algebra). The paper wrote the software that scans the barcode and instantly tells you exactly which book it is, even if the book is written in a language nobody speaks yet.

Why Does This Matter?

The author mentions that this work is a "prequel" to future papers.

The Future Goal: To understand Prismatic Cohomology better. This is a new way of doing math that might solve some of the hardest problems in number theory (like understanding the deep secrets of prime numbers).
The Impact: By building this toolbox now, the author is paving the road for future mathematicians to drive their cars (theories) over it without hitting potholes. Without this paper, trying to study these "fuzzy, zoomed-in, prismatic" shapes would be like trying to build a skyscraper with a hammer made of jelly. This paper gives them a steel hammer.

Summary in One Sentence

This paper builds a new, sturdy mathematical toolkit that allows us to manipulate complex, "fuzzy" geometric shapes when they are infinitely zoomed-in and filtered, proving that these shapes can be perfectly classified and understood using a clever "time-lapse" trick.

1. Problem Statement and Motivation

The paper addresses the interaction between derived algebras (in the sense of Bhatt-Mathew and Raksit) and formal derived geometry, specifically within the context of prismatic cohomology and prismatization.

The Gap: While the theory of derived rings (non-connective animated rings) is well-established for discrete rings and connective settings, its behavior in formal completions (specifically $J$ -adic completions) and on formal stacks was not rigorously developed in the literature.
The Obstruction: Standard definitions of derived algebraic contexts (e.g., Raksit's) often require the $t$ -structure to be compatible with filtered colimits. However, in the category of $p$ -complete modules ( $Mod_{\mathbb{Z}_p}^{\wedge}$ ), the coconnective part is not stable under filtered colimits (e.g., the colimit of $p^n$ -torsion modules can become connective). This prevents a direct application of existing monadicity theorems to define derived algebras on formal spectra like $Spf(\mathbb{Z}_p)$ .
The Goal: To develop a robust theory of $J$ -complete derived algebras and classify them on specific geometric objects (formal stacks) relevant to prismatic cohomology, such as the classifying stack $BG_m$ and filtered stacks like $\mathbb{A}^1/G_m$ .

2. Methodology

The author employs a combination of $\infty$ -categorical algebra, monadicity/comonadicity theorems, and descent theory.

$J$ -Complete Derived Algebras: The paper defines the $\infty$ -category $DAlg_A^{\wedge J}$ of $J$ -complete derived algebras as a localizing subcategory of $DAlg_A$ . It proves that this category is presentably symmetric monoidal and that the forgetful functor to $J$ -complete modules is monadic.
Formal Spectra via Kan Extensions: Instead of defining formal spectra ad-hoc, the author defines $Spf(A)$ as a prestack (the subfunctor of $Spec(A)$ annihilating powers of $J$ ) and defines derived algebras on it via right Kan extension. A key technical result is proving that $DAlg(Spf(A)) \simeq DAlg_A^{\wedge J}$ .
Comonadicity and Grading: The paper utilizes the Barr-Beck-Lurie theorem in a "continuous" setting. It establishes that graded derived algebras are comonadic over ungraded derived algebras via the action of a completed Hopf algebra (specifically $A[Z]$ for $G_m$ ).
Quasi-Affine Morphisms: The author extends the classification of derived algebras along quasi-affine morphisms (where $f: X \to Y$ is quasi-affine implies $DAlg(X) \simeq DAlg(Y)_{f_*\mathcal{O}_X/}$ ).
Rees Construction: The paper adapts the Rees construction (relating filtered objects to graded objects over a polynomial ring) to the derived setting, linking filtered derived algebras to graded derived algebras over $A[t]$ .

3. Key Contributions and Results

A. Theory of $J$ -Complete Derived Algebras

Theorem 1.10: Establishes that $DAlg_A^{\wedge J}$ is a presentably symmetric monoidal $\infty$ -category. The forgetful functor to $Mod_A^{\wedge J}$ is monadic with a left adjoint given by the $J$ -adic completion of the free derived algebra functor ( $LSym_A$ ).
Theorem 1.11: Proves a folklore result that $QCoh(Spf(A)) \simeq Mod_A^{\wedge J}$ , extending this equivalence to derived algebras: $DAlg(Spf(A)) \simeq DAlg_A^{\wedge J}$ .

B. Graded and Continuous Comonadicity

Theorem 1.12: Proves a "continuous" comonadic adjunction between graded $J$ -complete derived algebras and ungraded ones:
$(\pi!)^\wedge_J : GrDAlg_A^{\wedge J} \rightleftarrows DAlg_A^{\wedge J} : \pi^*$
The associated comonad is identified with the action of the completed bicommutative Hopf algebra $(A[Z])^\wedge_J$ .
Generalization: This result is extended to diagonalizable group schemes $D(M)$ , identifying graded algebras with comodules over $A[M]$ .

C. Classification on Geometric Stacks

Theorem 1.13: For an animated ring $A$ , there is an equivalence:
$DAlg(BG_m \times Spf(A)) \simeq GrDAlg_A^{\wedge J}$
This classifies derived algebras on the formal classifying stack of the multiplicative group.
Theorem 1.14: For a quasi-affine morphism $f: X \to Y$ , there is an equivalence $DAlg(X) \simeq DAlg(Y)_{f_*\mathcal{O}_X/}$ . This allows the classification of derived algebras on any stack that is quasi-affine over a base.

D. Filtered Stacks and Prismatic Gauges

Rees Equivalence: The paper constructs an equivalence between filtered derived algebras and graded derived algebras over the Rees ring:
$FilDAlg_A \simeq GrDAlg_{A[t]}$
Theorem 1.15 (Main Application): For an orientable bounded prism $(A, I)$ with generator $d \in I$ , there is an equivalence between filtered derived algebras (with specific completeness conditions) and derived algebras on a specific formal stack:
$FilDAlg_{Fil(d)A}^{(p,d)\text{-comp}} \simeq DAlg(Spf(A[u, t]/(ut - d))/G_m)$
Corollary 1.16: If $(A, I)$ is a perfect prism, this stack is isomorphic to the Nygaard filtered prismatization $(A/I)_{Nyg}$ , providing a classification of derived algebras on the Nygaard stack.

4. Significance and Impact

Foundational for Prismatic Cohomology: The results provide the necessary algebraic framework to treat prismatic gauges (objects in the theory of prismatic cohomology) as derived algebras on specific stacks. This allows for a rigorous treatment of coefficients in prismatic cohomology.
Tannakian Framework: The paper lays the groundwork for a Tannakian formalism in the formal derived setting, which is crucial for understanding the geometry of prismatic cohomology and its relation to $p$ -adic Hodge theory.
Resolution of Technical Obstructions: By explicitly handling the failure of $t$ -structure compatibility with filtered colimits in $p$ -complete settings, the paper resolves a major technical hurdle in applying derived algebraic geometry to $p$ -adic formal schemes.
Future Work: These results are foundational for the author's forthcoming work ([Sah26a], [Sah26b]), which aims to define relative Nygaard filtered prismatization and study syntomic cohomology and Frobenius liftable schemes using these stack-theoretic classifications.

In summary, the paper successfully bridges the gap between abstract derived algebraic geometry and the concrete needs of modern $p$ -adic geometry (prisms, Nygaard filtrations), providing a rigorous classification of derived algebras on formal stacks that were previously ill-defined in the literature.