Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

12.5%25.0%37.5%50.0% · Jan 199419922001200920172026
48 results for Modular Arithmetic

We investigate modular embeddings for semi-arithmetic Fuchsian groups. First we prove some purely algebro-geometric or even topological criteria for a regular map from a smooth complex curve to a quaternionic Shimura variety to be covered by a modular embedding. Then we set up an adelic formalism for modular embeddings…

2015-09-02abs ↗pdf ↗

Neural networks learn modular arithmetic but not all, extending known solutions to generalize.

problem Neural networks struggle with modular arithmetic, especially for polynomials.
method Developed analytical solutions for MLP networks to learn modular addition and multiplication, then combined these solutions to generalize on arbitrary modular polynomials.
result Neural networks can learn and generalize solutions to modular polynomials, supporting the hypothesis that some polynomials are learnable.

We determine the number of cusps of minimal Picard modular surfaces. The proof also counts cusps of other Picard modular surfaces of arithmetic interest. Consequently, for each N > 0 there are finitely many commensurability classes of nonuniform arithmetic lattices in SU(2, 1) that contain an N-cusped surface. We also …

2010-05-12abs ↗pdf ↗

Recursive Feature Machines show grokking in modular arithmetic without neural networks.

problem Grokking in modular arithmetic tasks.
method Recursive Feature Machines (RFM) with Average Gradient Outer Product (AGOP).
result RFM and neural networks learn block-circulant features to solve modular arithmetic.

Fuchsian groups with a modular embedding have the richest arithmetic properties among non-arithmetic Fuchsian groups. But they are very rare, all known examples being related either to triangle groups or to Teichmueller curves. In Part I of this paper we study the arithmetic properties of the modular embedding and deve…

2015-03-19abs ↗pdf ↗

We prove a rigidity theorem for semi-arithmetic Fuchsian groups: If Γ1Γ_1, Γ2Γ_2 are two semi-arithmetic lattices in PSL(2,R)\mathrm{PSL}(2,\mathbb{R}) virtually admitting modular embeddings and f ⁣:Γ1Γ2f\colonΓ_1\toΓ_2 is a group isomorphism that respects the notion of congruence subgroups, then ff is induced by an inner automor…

2014-08-13abs ↗pdf ↗

Transformers learn to solve modular arithmetic tasks by in-context learning and skill composition.

problem Understanding how large language models generalize to unseen tasks in modular arithmetic.
method Pre-training on a set of modular arithmetic tasks and evaluating out-of-distribution performance.
result Transformers require two transformer blocks for out-of-distribution generalization, and deeper models exhibit transient out-of-distribution performance.

The {\em Wiman-Edge pencil} is the universal family $\Cs/\mathcal B$ of projective, genus 66, complex-algebraic curves admitting a faithful action of the icosahedral group $\Af_5$. The goal of this paper is to prove that the monodromy of $\Cs/\mathcal B$ is commensurable with a Hilbert modular group; in particular is …

2019-11-04abs ↗pdf ↗

We show that the modular group has an infinite family of finite index subgroups, each of which has the same trace set as the modular group itself. Various congruence subgroups of the modular group, and the Bianchi groups, are also shown to have this property. In the case of the modular group, we construct examples of s…

2013-12-30abs ↗pdf ↗

Study explores how neural networks and Transformers learn modular arithmetic with multiple inputs.

problem Understanding how neural networks and Transformers learn modular arithmetic with multiple inputs.
method Analytical characterization of features learned by neural networks and Transformers, focusing on margin maximization and Fourier spectra.
result Neural networks and Transformers require a minimum neuron count of \( m \geq 2^{2k-2} \cdot (p-1) \) to solve modular addition problems with \( k \) inputs and modulus \( p \).

We survey contributions of Robert MacPherson to the theory of arithmetic groups. There are two main areas we discuss: (i) explicit reduction theory for Siegel modular threefolds, and (ii) constructions of compactifications of locally symmetric spaces. The former is joint work with Mark McConnell, the latter with Lizhen…

2006-03-15abs ↗pdf ↗

New geometric invariant limits the number of semi-arithmetic groups.

problem Understanding the structure of semi-arithmetic Fuchsian groups.
method Introducing a new geometric invariant called stretch and using the arithmetic Margulis lemma.
result There exist only finitely many conjugacy classes of semi-arithmetic groups with bounded arithmetic dimension, stretch, and coarea.

Study on coloring virtual tangles with integer and modular arithmetic.

problem Characterizing Fox colorings of virtual tangle diagrams.
method Analyzed classical and virtual tangle diagrams using vector representations and divisibility conditions.
result For R=ZR=\mathbb{Z}, realizability depends on divisibility of the alternating sum. For R=Z/pZR=\mathbb{Z}/p\mathbb{Z}, all vectors are realizable.

We show that the conjectural cusped complex hyperbolic 2-orbifolds of minimal volume are the two smallest arithmetic complex hyperbolic 2-orbifolds. We then show that every arithmetic cusped complex hyperbolic 2-manifold of minimal volume covers one of these two orbifolds. We also give all minimal volume manifolds that…

2010-05-24abs ↗pdf ↗

Study on scattering geodesics on modular surface and their sojourn times.

problem Distribution of scattering geodesics and their sojourn times on modular surface.
method Analysis of scattering geodesics in modular surface, establishing connection to prime divisors in arithmetic progression.
result Established a connection between scattering geodesics and prime divisors in arithmetic progression.

Sharp lower bounds for modular invariants and Dehn twist coefficients in genus 2 and 3.

problem Finding sharp lower bounds for modular invariants and Dehn twist coefficients.
method Analyzing the relation between fractional Dehn twists and modular invariants, classifying pseudo-periodic maps, and proving rigidity properties.
result Sharp lower bounds for modular invariants and Dehn twist coefficients in genus 2 and 3.

The study finds infinitely many semi-arithmetic Riemann surfaces with dense systoles and distinct invariant trace fields.

problem Existence and properties of semi-arithmetic Riemann surfaces.
method Combining number theory and hyperbolic geometry to prove existence and properties of semi-arithmetic Riemann surfaces.
result Existence of infinitely many semi-arithmetic Riemann surfaces with dense systoles and distinct invariant trace fields.

Holomorphic networks on modular arithmetic show clear success or failure, no in-between.

problem Understanding when neural networks can represent modular arithmetic tasks.
method Two-layer networks with holomorphic monomial activations trained on modular tasks.
result The network's output is confined to a subspace of characters, and representability depends on the task's Fourier support.

Mark and Paupert devised a general method for obtaining presentations for arithmetic non-cocompact lattices, ΓΓ, in isometry groups of negatively curved symmetric spaces. The method involves a classical theorem of Macbeath applied to a ΓΓ-invariant covering by horoballs of the negatively curved symmetric space upon w…

2019-08-31abs ↗pdf ↗

We present a general method to compute a presentation for any cusped arithmetic hyperbolic lattice ΓΓ, applying a classical result of Macbeath to a suitable ΓΓ-invariant horoball cover of the corresponding symmetric space. As applications we compute presentations for the Picard modular groups ${\rm PU}(2,1,\mathcal{O…

2017-09-20abs ↗pdf ↗

Over-parameterized models can memorize noisy labels and still generalize well, revealing a hidden structure.

problem Understanding how over-parameterized models can simultaneously memorize noisy labels and generalize well.
method Investigated through modular arithmetic tasks with label noise using two-layer neural networks.
result Over-parameterized models can achieve near-perfect test accuracy with 80% label noise by extracting an internal generalization structure.

Persistent homology reveals a topological signature of grokking in neural networks.

problem Understanding how neural networks learn and generalize from modular arithmetic tasks.
method Persistent homology on point clouds derived from embedding matrices of models trained on modular arithmetic.
result A sharp increase in first homology persistence indicates grokking, with a dominant long-lived topological feature and structured secondary features.

The f-invariant is a higher version of the e-invariant that takes values in the divided congruences between modular forms; it can be formulated as an elliptic genus of manifolds with corners of codimension two. In this thesis, we develop a geometrical interpretation of the f-invariant in terms of index theory, thereby …

2008-08-04abs ↗pdf ↗

Linking numbers of modular knots derived from geometric and algebraic properties.

problem Understanding linking numbers between modular knots and the trefoil.
method Geometric and algebraic properties of the modular group and its action on the hyperbolic plane.
result Derived several formulae for linking numbers with arithmetical, combinatorial, topological and group theoretical flavors.

Develops Weil bundles over \( p \)-adic manifolds for arithmetic geometry.

problem Connecting differential calculus and arithmetic geometry over \( p \)-adic fields.
method Systematic theory of Weil bundles, developing analytic structures.
result Establishes canonical analytic structures on Weil bundles and their cohomological comparison.

The Quantum Modularity Conjecture of Zagier predicts the existence of a formal power series with arithmetically interesting coefficients that appears in the asymptotics of the Kashaev invariant at each root of unity. Our goal is to construct a power series from a Neumann-Zagier datum (i.e., an ideal triangulation of th…

2015-11-18abs ↗pdf ↗

Study shows how transformers learn to combine simple tasks into complex ones.

problem Understanding how transformers learn to perform complex tasks not seen during training.
method Controlled setting involving variable assignment and modular addition; partitioned training data analysis.
result Small transformers can generalize to unseen combinations of variables and numbers.

We study the covolumes of arithmetic lattices in PSL2(R)nPSL_2(\mathbb R)^n for n2n\geq 2 and identify uniform and non-uniform irreducible lattices of minimal covolume. More precisely, let μμ be the Euler-Poincaré measure on PSL2(R)nPSL_2(\mathbb R)^n and χ=μ/2nχ=μ/2^n. We show that the Hilbert modular group $PSL_2(\mathfrak o_{k_{49…

2015-01-26abs ↗pdf ↗