--- title: Artin–Mazur Formal Group Functors url: https://www.emergentmind.com/topics/artin-mazur-formal-group-functors type: topic --- # Artin–Mazur Formal Group Functors Artin–Mazur formal group functors are contravariant deformation-theoretic functors extracted from higher \(\mathbf G_m\)-cohomology of a smooth proper scheme. In their classical form, for a smooth proper variety \(X\) over a field or over a suitable base ring, they assign to a local Artinian algebra the kernel of the restriction map in \(H^i(-,\mathbf G_m)\); under mild hypotheses they are pro-representable by commutative formal Lie groups, and their tangent spaces are canonically identified with \(H^i(X,\mathcal O_X)\) [1605.06440]. More recent work places these functors simultaneously in fppf, Witt-vector, crystalline, and unipotent-homotopical frameworks, yielding criteria for formal smoothness, explicit coordinate realizations for hypersurfaces, and a reconstruction of the usual Artin–Mazur formal groups without ad hoc cohomology-vanishing assumptions [2510.03001; 2510.06152]. ## 1. Definition and cohomological meaning Let \(X\) be a smooth proper scheme over a Noetherian ring \(R\) of characteristic \(p>0\), or more generally over a complete discrete valuation ring of mixed characteristic \((0,p)\). The \(i\)-th Artin–Mazur formal group functor is defined on local Artinian \(R\)-algebras \(A\) by \[ \Phi^i_X(A)=\Ker\!\Bigl[ H^i(X\times_R A,\mathbf G_m)\to H^i(X,\mathbf G_m) \Bigr]. \] Over an algebraically closed field \(k\) of characteristic \(p>0\), the fppf formulation used in later work is \[ \Phi^i(X,\mathbf G_m)(R) = \ker\Bigl\{ H^i_{\mathrm{fppf}}(X_R,\mathbf G_m) \longrightarrow H^i_{\mathrm{fppf}}(X_{R_{\mathrm{red}}},\mathbf G_m) \Bigr\}. \] After fppf-sheafification, these functors are representable by formal groups of finite type over \(k\); when they are representable one recovers the original functor. Via the Kummer sequence \(\mu_{p^n}\to \mathbf G_m\to \mathbf G_m\), one also identifies \(\Phi^i(X,\mathbf G_m)\) with the formal completion of \[ R^if_*\,\mu_{p^\infty}=\varinjlim_n R^if_*\,\mu_{p^n} \] at the identity section [2510.03001]. The basic linearization is provided by the tangent space. Under mild hypotheses, \(\Phi^i_X\) is pro-representable by a commutative formal Lie group whose tangent space is canonically identified with \(H^i(X,\mathcal O_X)\). In hypersurface situations this dimension can be computed explicitly from the interior lattice points of the Newton polytope. This places the functor at the interface of infinitesimal deformation theory, crystalline methods, and \(p\)-adic invariants. The theory generalizes the classical formal Picard group and the formal Brauer group and is tied, in the formulation of Vlasenko, to \(p\)-adic variation of zeta- and \(L\)-functions, period maps, and deformation theory of algebraic cycles [1605.06440]. ## 2. Representability, formal smoothness, and cohomological criteria Formal smoothness is not automatic once representability is known. A precise criterion is given by Grammatica through the auxiliary notion of \(C\)-smoothness, where \(C=k[\mathbf Q_p/\mathbf Z_p]\) and \(f_C:C\to C\) is induced by multiplication by \(p\) on \(\mathbf Q_p/\mathbf Z_p\). For a presheaf of abelian groups \(G\), one says that \(G\) is \(C\)-smooth when \(G(f_C):G(C)\to G(C)\) is surjective. If \(G\) is a formal group of finite type, \(C\)-smoothness agrees with ordinary formal smoothness. In particular, \(\Phi^i(X,\mathbf G_m)\) is formally smooth if and only if \(\varinjlim_n R^if_*\,\mu_{p^n}\) is \(C\)-smooth, equivalently if and only if \[ (\mathrm{id}_X\times f_C): H^i(X_C,\mathbf Q_p/\mathbf Z_p(1)) \longrightarrow H^i(X_C,\mathbf Q_p/\mathbf Z_p(1)) \] is surjective [2510.03001]. The same paper relates formal smoothness to torsion phenomena in \(p\)-adic cohomology. One has the split exact sequence \[ 0\to H^i(X_C,\mathbf Z_p(1))\otimes \mathbf Q_p/\mathbf Z_p \longrightarrow H^i(X_C,\mathbf Q_p/\mathbf Z_p(1)) \longrightarrow H^{i+1}(X_C,\mathbf Z_p(1))_{\mathrm{tors}} \to 0, \] so smoothness is controlled by surjectivity and torsion in adjacent degrees. If \(\Phi^i(X,\mathbf G_m)\) is representable and \(H^{i+1}(X/W(k))\) is \(p\)-torsion-free in crystalline cohomology, then \(\Phi^i(X,\mathbf G_m)\) is formally smooth. Conversely, if \(\Phi^i(X,\mathbf G_m)\) is representable but \(H^{i+1}(X,\mathbf Z_p)\) has non-zero torsion, then \(\Phi^i(X,\mathbf G_m)\) is not formally smooth. For abelian varieties, torsion-freeness of crystalline cohomology yields formal smoothness of all \(\Phi^i(A,\mathbf G_m)\) [2510.03001]. A further reformulation uses Serre’s Witt-vector cohomology. If \(\Phi^i(X,\mathbf G_m)\) is representable, then the following are equivalent: \(\Phi^i(X,\mathbf G_m)\) is formally smooth; the natural map \[ H^i(X,W)\to H^i(X,\mathcal O_X) \] is surjective; and the Verschiebung \(V\) on \(H^{i+1}(X,W)\) is injective. This equivalence makes formal smoothness a concrete Witt-cohomological condition rather than merely a property of a functor. It also shows that representability and formal smoothness should be kept distinct: the paper constructs, for every \(d\ge 2\), smooth proper varieties \(X\) for which \(\Phi^i(X,\mathbf G_m)\) is formally smooth for \(i0. \] For each \(m\ge 1\), define the \(g\times g\) matrix \(\beta_m\) by \[ \beta_{m,u,v}=\text{ coefficient of }x^{mv-u}\text{ in }f(x)^{m-1}, \qquad \beta_1=I_g. \] The logarithm series is \[ l(\tau)=\sum_{m=1}^\infty \beta_m \tau^m/m, \] and the associated formal group law over \(R\otimes \mathbf Q\) is \[ G_f(\tau,\tau')=l^{-1}(l(\tau)+l(\tau')). \] Vlasenko proves that in fact \[ G_f(\tau,\tau')\in R_{(p)}[[\tau,\tau']]^g, \] so \(G_f\) is an integral \(g\)-dimensional formal group law over \(R_{(p)}\). If \(f\) is homogeneous of degree \(d>n\) and the corresponding projective hypersurface \(X_f\) is smooth over \(R\), then \(G_f\) is precisely the coordinate expression of the Artin–Mazur formal group functor \(\Phi^{n-1}_{X_f}\) [1605.06440]. This coordinate description interacts with higher Hasse–Witt matrices. Writing \(\alpha_s=\beta_{p^s}\), the reductions \(\bar\alpha_s\) control \(p\)-adic congruences, and when \(\bar\alpha_1\) is invertible the limits \[ F_\sigma=\lim_{s\to\infty}\alpha_{s+1}\sigma(\alpha_s)^{-1}, \qquad N_D=-\lim_{s\to\infty} D(\alpha_s)\alpha_s^{-1} \] exist \(p\)-adically. Their conjectural meaning is that they coincide with the Frobenius and Gauss–Manin connection matrices on the unit-root crystal \(H^{n-1}_{\mathrm{unit}}\), hence encode the crystal dual to the Artin–Mazur formal group [1605.06440]. In the one-dimensional case, Vlasenko gives a separate integrality theory for formal group laws in terms of the coefficients of the strict logarithm \[ f(x)=\sum_{n=1}^\infty \frac{b_{n-1}}{n}x^n \] and the canonical invariant differential \[ \omega=f'(x)\,dx=\Bigl(\sum_{n=0}^\infty b_n x^n\Bigr)\,dx. \] The paper gives a criterion of integrality of a one-dimensional formal group law in terms of congruences satisfied by these coefficients, and a \(p\)-adic analytic formula for the local characteristic polynomial at \(p\). Applied to Artin–Mazur formal groups of hypersurfaces with exactly one interior lattice point, this yields explicit logarithms \[ \log_{F_X}(x)=\sum_{n=0}^\infty b_n \frac{x^{n+1}}{n+1} \] with \[ b_n= \bigl[x_0^{w_0n}x_1^{w_1n}\cdots x_m^{w_mn}\bigr]\,V(x_0,\dots,x_m)^n, \] and proves \(p\)-integrality under the stated denominator condition on the coefficients of \(V\) [1509.06002]. ## 4. Unipotent spectra and reconstruction of the formal groups A major reformulation comes from the theory of unipotent spectra. Fix a commutative ring \(A\), and let \(\AffSt_{A*}\) be Toën’s \(\infty\)-category of pointed affine stacks over \(A\), equipped with the loop endofunctor \(\Omega X=* \times_X *\). The \(\infty\)-category of unipotent spectra is defined by \[ \Sp_A^U = \lim\!\Bigl( \cdots \xrightarrow{\Omega}\AffSt_{A*}\xrightarrow{\Omega}\AffSt_{A*} \Bigr) \simeq \Sp(\AffSt_{A*}), \] the stabilization of pointed affine stacks. There is a left adjoint \(\Sigma^\infty_+\) and a right adjoint \(\Omega^\infty\); for a stack \(Y\), the object \(\Sigma^\infty_+Y\in \Sp_A^U\) is its unipotent stable homotopy type. Its homotopy sheaves \(\pi_i(\Sigma^\infty_+Y)\) are representable by commutative unipotent affine group schemes, called the unipotent stable homotopy groups of \(Y\) [2510.06152]. Passing to \(\mathbf Z\)-linear objects over a field \(k\), one obtains unipotent homology \[ H_*^U(Y)=\Sigma^\infty_+Y\otimes_{\mathbf S}\mathbf Z. \] For a finite-dimensional \(k\)-scheme \(X\), this admits an increasing coniveau filtration \[ F^pH_*^U(X) = \lim_{\substack{Z\subset X\\ \mathrm{codim}\,Z\ge p}} H_*^U(X\setminus Z), \qquad 0=F^{n+1}\subset \cdots \subset F^0=H_*^U(X), \] with associated graded \[ \mathrm{gr}^pH_*^U(X)\simeq \prod_{x\in X^{(p)}} H^U_{*,x}(X_x), \] and a convergent homological spectral sequence \[ E^1_{p,q} = \prod_{x\in X^{(p)}} H^U_{p+q,x}(X_x) \Longrightarrow H^U_{p+q}(X). \] The local terms are related to flat cohomology by \[ \RHom\bigl(H^U_{*,x}(X_x),G\bigr)\simeq R\Gamma_x(X,G) \] for every commutative unipotent \(G\). This comparison yields exact sequences linking unipotent local homology to Witt-vector cohomology and sets up the reconstruction of the Artin–Mazur functors [2510.06152]. The main reconstruction theorem states that over any perfect field \(k\) of characteristic \(p>0\), for every smooth proper \(X/k\) and each \(p\ge 0\), the Cartier dual of the fppf-sheafified Artin–Mazur functor is canonically isomorphic to the term \(E_2^{p,0}\) on the second page of the coniveau spectral sequence: \[ \bigl((\Phi_X^p)^{fl}\bigr)^\vee \cong E_2^{p,0} = \ker\Bigl( \mathrm{gr}^pH_*^U(X)\xrightarrow{d^1}\mathrm{gr}^{p-1}H_*^U(X) \Bigr). \] Accordingly, \[ F_X(A) = \Hom\bigl((\Phi_X^p)^{fl,\vee},W(A)\bigr) = \Hom\bigl(E_2^{p,0},\Spec W(A)\bigr) \] is the formal group functor arising from unipotent stable homotopy. The conceptual point is that the usual Artin–Mazur formal groups are recovered without any vanishing assumptions on the intermediate \(\mathcal O_X\)-cohomology [2510.06152]. ## 5. Heights, Witt cohomology, and quasi-\(F\)-split geometry When the Artin–Mazur functor is one-dimensional, its height provides a refined measure of its \(p\)-typical structure. Let \(\kappa\) be a perfect field of characteristic \(p>0\), and let \(X\) be a proper log-smooth scheme of Cartier type over \(\kappa\). For each \(q\ge 0\), the Artin–Mazur functor \[ \Phi^q_{X/\kappa}(A) = \ker\Bigl( H^q_{\mathrm{\acute et}}(X_A,\mathbf G_m) \to H^q_{\mathrm{\acute et}}(X,\mathbf G_m) \Bigr) \] has height \[ h^q_{\mathrm{AM}}(X/\kappa) = \min\Bigl\{ n\ge 1 \,\Bigm|\, F:H^q(X,W_n\mathcal O_X)\to H^q(X,W\mathcal O_X) \text{ is nonzero} \Bigr\}, \] with the convention \(h^q_{\mathrm{AM}}=\infty\) if all such Frobenius maps vanish [1902.00185]. Nakkajima relates this invariant to Yobuko’s height \[ h_F(X) = \min\Bigl\{ n\ge 1 \,\Bigm|\, \exists\,\rho:F_*W_n(\mathcal O_X)\to \mathcal O_X \text{ with }\rho\circ F=R \Bigr\}, \] where \(R:W_n(\mathcal O_X)\to \mathcal O_X\) is the truncation map. The condition \(h_F(X)<\infty\) is called quasi-\(F\)-splitness. Under the stated hypotheses—one-dimensional \(H^q(X,\mathcal O_X)\), vanishing \(H^{q+1}(X,\mathcal O_X)=0\), vanishing Bockstein operators, and pro-representability of \(\Phi^q_{X/\kappa}\)—one has the fundamental inequality \[ h^q_{\mathrm{AM}}(X/\kappa)\le h_F(X). \] This exhibits Artin–Mazur height as bounded above by a splitting invariant defined directly on Witt sheaves [1902.00185]. The same framework yields finiteness statements for Witt cohomology. If \(X\) is quasi-\(F\)-split, then for every \(i,j\ge 0\), \[ H^j(X,W\Omega^i_{X/s}) \] is a finitely generated \(W(\kappa)\)-module; in particular \(H^j(X,W\mathcal O_X)\) is finitely generated for every \(j\). For a proper smooth threefold \(Y/\kappa\), if \(H^2(Y,W\mathcal O_Y)\) is finitely generated over \(W(\kappa)\), then the \(p\)-primary torsion subgroup \(\CH^2(Y)\{p\}\) is of finite cotype. Combining the results, a quasi-\(F\)-split proper smooth threefold has \(\CH^2(Y)\{p\}\) of finite cotype. In the Calabi–Yau threefold case, finite third Artin–Mazur height implies the same conclusion through the equality \(h_F(Y)=h_{\mathrm{AM}}^3(Y)\) noted in the paper [1902.00185]. ## 6. Comparisons, standard examples, and duality refinements Several standard examples clarify the range of the theory. In degree \(p=1\), the unipotent-spectral reconstruction gives \[ E_2^{1,0}\simeq (\Pic_X^0)^{\wedge p}, \qquad \Phi_X^1=\widehat{\Pic_X^0}. \] This is the Artin–Mazur formal Picard group. In degree \(p=2\) for surfaces, the same formalism recovers the Brauer-type example. Under the classical connectivity assumptions \[ H^i(X,\mathcal O_X)=0\qquad (0