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.

169,341 papers · 148 categories

Trend · papers per month

76152228304 · Jun 202019922001200920182026
48 results for stochastic copies

SmoothOut improves deep learning by smoothing out sharp minima, enhancing generalization.

problem Sharp minima in deep neural networks lead to poor generalization.
method SmoothOut framework that perturbs multiple copies of the DNN by noise injection and averages them, improving generalization.
result SmoothOut eliminates sharp minima and improves generalization in both small-batch and large-batch training.

Study on how deterministic dependencies affect information synergy and redundancy.

problem Understanding how deterministic dependencies impact information synergy and redundancy.
method Systematic analysis of deterministic dependencies on information decomposition.
result Identifies how negative terms can originate from deterministic dependencies and discusses implications for neural coding.

This paper introduces DPI, a new metric to assess data-copying risk in tabular data.

problem Measuring privacy risk of data-copying in tabular generative models.
method Proposes Data Plagiarism Index (DPI) for evaluating data-copying risk.
result DPI identifies data-copying threats to tabular data models, highlighting privacy and fairness issues.

New seq2seq model can copy entire spans, outperforming simpler models in editing tasks.

problem Editing documents or source code using seq2seq models with explicit token copying.
method Extended seq2seq model capable of copying entire input spans to output in one step, new training and inference methods.
result New model consistently outperforms simpler baselines in editing tasks of natural language and source code.

Paper analyzes gradient descent with noisy data copies for linear regression, showing regularization and acceleration effects.

problem Improving generalization in machine learning through data augmentation with noise.
method Gradient descent with on-line noisy copies for linear regression analysis.
result Training with on-line noisy copies is equivalent to ridge regularization with a specific regularization parameter.

Shapes can roll downhill following any curve, but often return to initial orientation after crossing multiple copies.

problem How to design shapes that roll downhill along a given curve and its translations.
method Analyzing the geometric properties and motion of shapes on inclined planes.
result Most curves allow shapes to roll downhill following them and their translations, but some require crossing multiple copies.

Paper proposes a neural network for generating better questions from text.

problem Automatic generation of relevant questions from sentences and paragraphs.
method Adaptive copying recurrent neural network model with a copying mechanism added to a bidirectional LSTM architecture.
result The model outperforms state-of-the-art methods in question generation metrics.

GMC benchmark isolates retrieval in Transformers, revealing max-margin alignment.

problem Understanding how Transformers develop match-and-copy behavior on natural data.
method Introducing Gaussian Match-and-Copy (GMC) as a minimalist benchmark.
result Gradient descent drives parameters to diverge while aligning with max-margin separator.

Bayesian theory explains abrupt emergence of copy subcircuit in attention.

problem Understanding the abrupt emergence of the copy subcircuit in attention during training.
method Deriving a closed-form posterior over the attention matrix and reducing it to a low-dimensional order parameter space.
result Derive a phase transition in the amount of training data.

Under-parameterized networks can either copy or average teacher weights, leading to universal optimal solutions.

problem Approximating a teacher network with an under-parameterized student network.
method Analyzing shallow neural networks with erf activation function and unitary teacher weights, proving copy-average configurations are critical points and finding the optimal solution.
result The optimal solution for under-parameterized networks has a universal structure, whether copying or averaging teacher neurons.

The paper introduces a new method to improve model generalization by routing model copies through permutations.

problem Improving model generalization in machine learning.
method The method replicates a model \(M\) times and rewire the contexts in which local learning messages are computed using permutations.
result The method improves generalization by structured message sharing rather than coupling parameters.

A simple patch copying method reduces black-box adversarial attack queries by 81%.

problem The effectiveness of black-box adversarial attacks depends on the initialization method.
method Copying small patches from other images as initialization points.
result Reduces the number of queries required for a state-of-the-art Boundary Attack by 81%

A scalable algorithm for sampling and fine-tuning models using Tilt Matching.

problem Efficient sampling and fine-tuning of generative models.
method Tilt Matching, arising from a dynamical equation, minimizes variance and inherits regularity from stochastic interpolants.
result Empirically verified to be efficient and highly scalable, providing state-of-the-art results.

A free action of the direct product of two copies of the symmetric group on 3 elements on the cartesian product of two copies of the 3-sphere is constructed. This nonlinear action is constructed using surgery. The action provides a counterexample to a conjecture of Lewis made in 1968.

1998-06-06abs ↗pdf ↗

A fibration of Rn{\mathbb R}^n by oriented copies of Rp{\mathbb R}^p is called skew if no two fibers intersect nor contain parallel directions. Conditions on pp and nn for the existence of such a fibration were given by Ovsienko and Tabachnikov. A classification of smooth fibrations of R3{\mathbb R}^3 by skew oriente…

2014-12-29abs ↗pdf ↗

For a closed 4-manifold X and closed 3-manifold M we investigate the smallest integer n (perhaps infinity) such that M embeds in the connected sum of n copies of X. It is proven that any lens space (or homology lens space) embeds topologically locally flatly in a connected sum of 8 copies of the complex projective plan…

2001-12-21abs ↗pdf ↗

We construct branched double coverings by certain direct products of manifolds for connected sums of copies of sphere bundles over the 2-sphere. As an application we answer a question of Kotschick and Loeh up to dimension five. More precisely, we show that: (1) every simply connected, closed four-manifold admits a bran…

2012-10-04abs ↗pdf ↗

We discuss the relationship between the m-th homotopy group of the one-point union of r copies of the two-dimensional sphere and the m-th homotopy group of the one-point union of r+1 copies of the Thom space of the oriented two-dimensional universal vector bundle. Using a suitably choosen isomorphism between them a for…

2002-05-28abs ↗pdf ↗

Characterizes a specific type of Courant algebroid with a Calabi-Yau structure.

problem Understanding specific types of Courant algebroids with Calabi-Yau structures.
method Explains how a homotopy BV algebra with certain properties characterizes these algebroids.
result A Courant algebroid with a Calabi-Yau structure is a homotopy BV algebra with specific properties.

End-to-end character-level model for text generation without delexicalization.

problem Generating text without delexicalization and tokenization.
method Character-level sequence-to-sequence model with attention mechanism, copy mechanism, and transfer learning.
result Competitive performance in text generation metrics.

CORE optimizes molecules by copying or generating substructures, improving accuracy.

problem Inaccurate substructure prediction in molecule optimization.
method Copy & Refine (CORE) strategy combining scaffolding tree generation and adversarial training.
result Significant improvement in various molecule optimization metrics.

A new model for graph sampling that preserves structure without explicit targeting.

problem Graphs are often not fully representative of true relationships, leading to biased machine learning models.
method Node copying model: randomly replaces each node's neighbors with those of a randomly sampled similar node.
result The model achieves higher accuracy in node classification and mitigates adversarial attacks.

CACTI improves tabular data imputation by leveraging missingness patterns and contextual information.

problem Tabular data imputation with improved accuracy and robustness.
method Masked autoencoding approach with median truncated copy masking and contextual information.
result Average R2R^2 gain of 7.8% over the next best method across various datasets and missingness conditions.

Copycat CNN learns from random data to mimic target models.

problem Protecting state-of-the-art CNNs from copying.
method Querying target CNN with random non-labeled data to create a fake dataset, then training a copycat network.
result Copycat CNNs can achieve similar performance to target models, even from non-problem domain data.

Zero-Copy Architecture Detects Cross-Company Financial Signals Instantly.

problem Financial models miss cross-company disruptions due to static data.
method Heterogeneous Rust-Python streaming architecture that maps cross-company attention as a continuous-time graph.
result Zero-copy parsing and inference process delivers real-time cross-company signal detection.