Abstract:We prove that the space of algebraic maps between two smooth projective varieties, under certain conditions, admit a configuration space model, thereby obtaining an algebro-geometric analogue of Bendersky-Gitler's result on topological function spaces. Our result is a natural higher dimensional counterpart of \cite[Theorem 3]{Ban24}.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
This paper aims to solve a key problem in algebraic geometry, that is, under specific conditions, whether the space of algebraic maps between two smooth projective varieties can have a configuration - space model. Specifically, the author proves that under certain conditions, the space of algebraic maps between two smooth projective varieties can be described by a configuration - space model, thereby establishing an algebraic - geometry version of the Bendersky - Gitler theorem ([BG91, Theorem 7.1]), which was originally about topological function spaces.
#### Main research content:
1. **Background and motivation**:
- Research on continuous - map spaces and their connection with configuration spaces has a long history, involving the work of Anderson, Bendersky - Gitler, Snaith, Cohen - May - Taylor, etc.
- These studies show the stable splitting of function spaces under certain conditions, where the components of the splitting include configuration spaces.
2. **Main results**:
- The author proves that under certain conditions, the space of algebraic maps between two smooth projective varieties can have a configuration - space model, which provides a new perspective for algebraic geometry.
- Specifically, let \(X\) be an \(n\)-dimensional smooth projective variety, \((Y, \Upsilon)\) be a polarized smooth projective variety, and \(d\in N^1(X)\) be an acyclic class, then there exists a first - quadrant spectral sequence:
\[
E^{p,q}_1\Rightarrow H^{p + q}_c(\text{Mor}_d(X,Y);\mathbb{Q})
\]
where
\[
E^{p,*}_1=(H_*(\dot{X}^p;\mathbb{Q})\otimes \text{sgn}_{S_p})^{S_p}\otimes H_*(\text{Pic}_d(X);\mathbb{Q})\otimes H^*_c(Y(D_{p - 1});\mathbb{Q})
\]
for all \(p\leq r(d)+ 1\), here \(Y(D_p)\) are some auxiliary schemes that satisfy the conditions in Section 2.5.4, and \(\dot{X}^p := X^p-\{\text{diagonal}\}\).
3. **Important formulas and concepts**:
- Definition of the numerical class \(d\in N^1(X)\) separating \(r\) points:
\[
r(d):=\max\left\{r:d\text{ separates }r\text{ points}\right\}
\]
- When \(\delta := d - c_1(K_X)\) is ample, the stable bound \(r(d)\) satisfies:
\[
r(d)=\left\lfloor\min_{[W]\in CH_k(X),1\leq k\leq n}\frac{(\delta^k\cdot [W])}{k-\frac{n}{2}+n - 1}\right\rfloor - 1
\]
4. **Methodology**:
- The author uses sheaf - theoretic methods, especially by constructing Poincaré - dual Koszul - type chain complexes through appropriate compactifications and hyper - coverings.
- Tools in modern algebraic geometry, such as higher - category theory and the six - functor formalism of constructible sheaves, are also utilized in the proof process.
### Summary
The main contribution of this paper is to generalize the classical results of topological function spaces to the field of algebraic geometry and establish the connection between algebraic - map spaces and configuration spaces. This achievement not only deepens our understanding of algebraic geometry but also provides new tools and perspectives for further research.