Relative Mirror Symmetry for Non-Fano Varieties

Basic Information

• Paper ID: 2510.13027
• Title: Relative Mirror Symmetry for Non-Fano Varieties
• Author: Fenglong You (University of Nottingham)
• Classification: math.AG (Algebraic Geometry)
• Publication Date: October 16, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.13027

Abstract

This paper investigates mirror symmetry for logarithmic Calabi-Yau pairs $(X,D)$ consisting of a smooth projective variety $X$ and its smooth anticanonical divisor $D$, without assuming that $D$ is nef. The author considers the mirror proper Landau-Ginzburg model $(\check{X},W)$ arising from the Gross-Siebert intrinsic mirror construction, and studies the relationship between the normalized quantum period of $X$ and the classical period of $W$. The paper identifies that the discrepancy between them originates from curve counting in $D$, controlled by the mirror map associated with $D$. An explicit formula for the proper potential function $W$ encoding this discrepancy is obtained. Finally, it is proven that the quantum period together with the mirror map provides exactly the same information as the proper potential function.

Research Background and Motivation

Problem Background

1. Limitations of Classical Mirror Symmetry: Traditional mirror symmetry theory primarily focuses on Fano varieties, where the anticanonical divisor $-K_X$ is nef. However, many important geometric objects do not satisfy this condition.
2. Need for Tyurin Degenerations: When studying Tyurin degenerations of Calabi-Yau varieties, two logarithmic Calabi-Yau pairs $(X_1,D)$ and $(X_2,D)$ appear, where $-K_{X_1}$ and $-K_{X_2}$ are not necessarily nef. These varieties are called quasi-Fano varieties.
3. Concrete Example: An important example is the Tyurin degeneration of a smooth quintic threefold $Q_5$: $Q_5 \rightsquigarrow Q_4 \cup_{K3} \text{Bl}_C \mathbb{P}^3$ where $\text{Bl}_C \mathbb{P}^3$ is the blowup of $\mathbb{P}^3$ along a complete intersection $C$, which is not Fano.

Research Motivation

1. Theoretical Completeness: To understand mirror symmetry for Tyurin degenerations, one needs to study relative mirror symmetry for logarithmic Calabi-Yau pairs where $-K_{X_i}$ is not nef.
2. Technical Challenges: When $D$ is not nef, traditional mirror symmetry formulas no longer apply, requiring new theoretical frameworks to handle additional degenerate contributions.

Core Contributions

1. Generalization of Relative Mirror Theorem: Extends the relative mirror theorem to the case where $D$ is not nef, establishing explicit mirror formulas.
2. Identification of Discrepancy: Proves that when $D$ is not nef, the discrepancy between the normalized quantum period and the classical period originates from curve counting mapping into the divisor $D$.
3. Explicit Formula for Proper Potential Function: Provides an explicit expression for the proper potential function $W$ encoding the above discrepancy.
4. Establishment of Equivalence: Proves that the quantum period together with the mirror map provides exactly the same information as the proper potential function.
5. Exact Results in Dimension Restrictions: In dimensions 2 and 3, proves that the classical period still coincides with the normalized quantum period.

Methodology

Problem Formulation

Study mirror symmetry for logarithmic Calabi-Yau pairs $(X,D)$ consisting of a smooth projective variety $X$ and its smooth anticanonical divisor $D$, without assuming $D$ is nef. The goal is to understand the relationship between quantum period, classical period, and proper potential function.

Theoretical Framework

1. Relative Gromov-Witten Theory

The paper employs genus-zero relative Gromov-Witten theory with negative contact orders. For topological type $\Gamma = (0,m,\beta,\vec{k})$, where:

• $0$ denotes genus zero
• $m$ is the number of marked points
• $\beta \in NE(X)$ is the curve class
• $\vec{k} = (k_1,\ldots,k_m) \in \mathbb{Z}^m$ are contact orders satisfying $\sum_{i=1}^m k_i = D \cdot \beta$

The relative Gromov-Witten invariant is defined as: $\langle \prod_{i=1}^m \tau_{a_i}(\alpha_i) \rangle_{(X,D)}^\Gamma = \int_{c_\Gamma(X/D)} \prod_{i=1}^m \psi_i^{a_i} \text{ev}_i^* \alpha_i$

2. Decomposition of Mirror Map

When $D$ is not nef, two types of mirror maps appear:

Mirror Map in $D$: Arising from the relative-local model $(Y,D_0)$, where $Y = \mathbb{P}(O_D \oplus N_D)$ and $D_0$ is the zero divisor. The mirror map takes the form: $\tau_D(y,z) = \sum_{k \geq 0, \delta \in H^*(D), \beta \in NE(D)} \tau_{D,\delta,k,\beta}(\iota_* \delta) z^k y^\beta$

Relative Mirror Map: Given by the $z^0$ coefficient of the relative I-function: $\tau(y) = \sum_{i=1}^r p_i \log y_i + \sum_{\beta \in NE(X), d=D \cdot \beta \geq 2} \left[ \sum_{l \geq 0, \beta' \in NE(X)} \frac{1}{l!} \langle [\text{pt}] \psi^{d-2}, \tau_D, \ldots, \tau_D \rangle_{0,1+l,\beta'}^X y^{\beta'} \right] \frac{y^\beta}{(d-1)!} [1]_{-d}$

3. Relative I-Function

For smooth logarithmic Calabi-Yau pairs, the relative I-function is: $I_{(X,D)}(y, \tau_D(y,z), z) = \sum_{\beta \in NE(X)} J_{X,\beta}(\tau_D(y,z), z) y^\beta \frac{\prod_{a \leq D \cdot \beta}(D + az)}{\prod_{a \leq 0}(D + az)} \prod_{\text{if } D \cdot \beta > 0} \frac{1}{D + (D \cdot \beta)z} [1]_{-D \cdot \beta}$

Main Theorems

Theorem 1 (Nef Case)

When $X$ is a smooth projective variety and $D$ is a smooth anticanonical divisor that is nef, the classical period defined by the proper potential function $W$ coincides with the normalized quantum period of $X$.

Theorem 2 (General Case)

When $D$ is not necessarily nef, the classical period coincides with the normalized quantum period after applying the mirror map in $D$: $\sum_{n \geq 0} [W^n]\vartheta_0 = 1 + \sum_{d \geq 2} \sum_{\beta \in NE(X), D \cdot \beta = d} (D \cdot \beta)! \left[ \sum_{l \geq 0, \beta' \in NE(X)} \frac{1}{l!} \langle [\text{pt}] \psi^{d-2}, \tau_D, \ldots, \tau_D \rangle_{0,l+1,\beta'}^X q^{\beta'} \right] q^\beta$

Theorem 3 (Proper Potential Function Formula)

$W = x + \sum_{n=1}^\infty \sum_{\beta: D \cdot \beta = n+1} n \langle [1]_1, [\text{pt}]_n \rangle_{0,2,\beta}^{(X,D)} t^\beta x^{-n} = x \exp(g(y(q)))$ where $g(y)$ is defined by absolute invariants and the mirror map.

Technical Innovations

1. New Methods for Handling Non-Nef Cases

• Identifies additional contribution terms through degeneration analysis
• Establishes connections between these contributions and the mirror map in $D$

2. Identities for Relative Gromov-Witten Invariants

Generalizes identities from You24b to the case where $D$ is not nef, including:

• Computational formulas for negative contact order invariants
• Equivalence relations under different contact order configurations

3. Lagrange Inversion and Bell Polynomials

Uses Lagrange inversion theorem and Bell polynomial identities to prove the equivalence between quantum period and proper potential function.

Experimental Results

Exact Results Under Dimension Restrictions

Theorem (2-dimensional and 3-dimensional cases): When $X$ is a 2-dimensional or 3-dimensional smooth projective variety and $D$ is a smooth anticanonical divisor, the classical period still coincides with the normalized quantum period.

Proof Strategy: When $D$ is an elliptic curve or K3 surface, its genus-zero Gromov-Witten invariants vanish, so the mirror map in $D$ is trivial.

Concrete Example: Blowup of $\mathbb{P}^3$

Consider the blowup $\text{Bl}_C \mathbb{P}^3$ of $\mathbb{P}^3$ along a complete intersection $C$, where $C$ is the intersection of a quartic and quintic hypersurface.

The I-function is: $z q_1^{H/z} q_0^{h/z} \sum_{d_1,d_0 \geq 0} \left( \frac{\prod_{k=1}^{4d_1+d_0}(4H + h + kz)}{\prod_{k=1}^{d_1}(H + kz)^4 \prod_{k=1}^{d_0}(h + kz)} \right) \left( \prod_{\text{if } d_0 > d_1} \frac{1}{h - H + (d_0-d_1)z} \right) [1]_{-d_0+d_1} q_1^{d_1} q_0^{d_0}$

Through computation, the corresponding mirror map and period relations can be obtained.

Related Work

Mirror Symmetry Theory

• Gross-Siebert Program: Intrinsic mirror symmetry construction
• Mirror Symmetry for Fano Varieties: Classical theory established in CCG+13 and others
• Relative Mirror Symmetry: Work by FTY19, You25, and others

Gromov-Witten Theory

• Negative Contact Order Theory: Theoretical framework established in FWY20
• Relative Invariants: Various structural properties and computational methods

Application Background

• Tyurin Degenerations: Mirror symmetry research in DHT17
• Quasi-Fano Varieties: Geometric properties in non-Fano cases

Conclusions and Discussion

Main Conclusions

1. Successfully generalizes relative mirror symmetry theory to the non-Fano case
2. Identifies and quantifies the discrepancy between quantum and classical periods
3. Establishes explicit computational formulas for the proper potential function
4. Proves equivalence relations between different geometric quantities

Limitations

1. Technical Complexity: Computations in the non-nef case are significantly more complex, requiring consideration of additional degenerate contributions
2. Dimension Restrictions: While exact results exist in low dimensions, high-dimensional cases still require mirror map corrections
3. Computational Challenges: Computing the mirror map in $D$ may be difficult in practical applications

Future Directions

1. Generalization to More General Cases: Consider more complex geometric configurations
2. Computational Methods: Develop more efficient computational techniques
3. Applications: Apply these theoretical results to concrete geometric problems

In-Depth Evaluation

Strengths

1. Theoretical Innovation: Successfully addresses an important technical challenge in mirror symmetry theory
2. Systematic Methodology: Provides a complete theoretical framework and computational tools
3. Deep Results: Reveals profound connections between geometric quantities
4. Clear Exposition: Well-structured paper with sufficient technical details

Weaknesses

1. Computational Complexity: Practical computations may be extremely difficult
2. Limited Examples: Relatively few concrete computational examples
3. Generalizability: Generalizability of certain results requires further verification

Impact

1. Theoretical Contribution: Provides important generalization to mirror symmetry theory
2. Technical Value: Developed methods may apply to other related problems
3. Application Prospects: Provides new tools for studying degenerations of Calabi-Yau varieties

Applicable Scenarios

• Mirror symmetry research for quasi-Fano varieties
• Geometric analysis of Tyurin degenerations
• Applications of relative Gromov-Witten theory
• Study of logarithmic Calabi-Yau pairs

References

The paper cites 33 related references, primarily including:

• GS19 Gross-Siebert's intrinsic mirror symmetry work
• FWY20 Negative contact order relative Gromov-Witten theory
• CCG+13 Mirror symmetry for Fano manifolds
• You24b Author's previous work on proper potential functions