arXiv research

XL-Editor improves sentence post-editing using XLNet's variable-length insertion probability.

Models learn to represent edits from natural language and code.

The Wikimedia Foundation has recently observed that newly joining editors on Wikipedia are increasingly failing to integrate into the Wikipedia editors' community, i.e. the community is becoming increasingly harder to penetrate. To sustain healthy growth of the community, the Wikimedia Foundation aims to quantitatively…

The increasingly photorealistic sample quality of generative image models suggests their feasibility in applications beyond image generation. We present the Neural Photo Editor, an interface that leverages the power of generative neural networks to make large, semantically coherent changes to existing images. To tackle…

In recent years, the importance of deep learning has significantly increased in pattern recognition, computer vision, and artificial intelligence research, as well as in industry. However, despite the existence of multiple deep learning frameworks, there is a lack of comprehensible and easy-to-use high-level tools for …

We give a survey of our joint ongoing work with Ali Chamseddine, Slava Mukhanov and Walter van Suijlekom. We show how a problem purely motivated by "how geometry emerges from the quantum formalism" gives rise to a slightly noncommutative structure and a spectral model of gravity coupled with matter which fits with expe…

Paper identifies food types from Yelp photos using machine learning.

An ongoing challenge in the analysis of document collections is how to summarize content in terms of a set of inferred themes that can be interpreted substantively in terms of topics. The current practice of parametrizing the themes in terms of most frequent words limits interpretability by ignoring the differential us…

As the number of contributors to online peer-production systems grows, it becomes increasingly important to predict whether the edits that users make will eventually be beneficial to the project. Existing solutions either rely on a user reputation system or consist of a highly specialized predictor that is tailored to …

Econophysics uses computer simulations to study financial markets.

This article is the author's contribution to the volume "Problems on mapping class groups and related topics" which will be published in December 2006, with Benson Farb as Editor. Various individuals were invited by Farb to submit open problems which seemed interesting to them about surface mapping class groups. We sin…

This article was submitted to a volume under preparation, with Benson Farb as the editor, on the topic of open problems in surface mapping class groups. The braid group B_n is the mapping class group of an n-times punctured disk. The Iwahori-Hecke algebra H_n is a quotient of the braid group algebra of B_n by a quadrat…

The New Yorker publishes a weekly captionless cartoon. More than 5,000 readers submit captions for it. The editors select three of them and ask the readers to pick the funniest one. We describe an experiment that compares a dozen automatic methods for selecting the funniest caption. We show that negative sentiment, hum…

Responds to discussion on probabilistic integration in statistical computation.

This is the original manuscript dated March 9th 1983, typeset by the Editors for the Proceedings of the Midwest Geometry Conference 2007 held in memory of Thomas Branson. Stephen Paneitz passed away on September 1st 1983 while attending a conference in Clausthal and the manuscript was never published. For more than 20 …

ChainGAN improves GANs by sequentially enhancing generated samples.

GraphQ system uses GNNs to search for subgraph patterns in graphs.

Online knowledge repositories typically rely on their users or dedicated editors to evaluate the reliability of their content. These evaluations can be viewed as noisy measurements of both information reliability and information source trustworthiness. Can we leverage these noisy evaluations, often biased, to distill a…

Model removes objects from general scenes using weak supervision.

A framework that edits existing examples to generate structured outputs.

This short note is intended as a "Letter to the Editor" Perspective in order that it serves as a contribution, in view of reaching the physics community caring about rare events and scaling laws and unexpected findings, on a domain of wide interest: sport and money. It is apparent from the data reported and discussed b…

Study shows Wikipedia views correlate with digital currency market performance.

This study investigates the content of the published scientific literature in the fields of operations research and management science (OR/MS) since the early 1950s. Our study is based on 80,757 published journal abstracts from 37 of the leading OR/MS journals. We have developed a topic model, using Latent Dirichlet Al…

The paper teaches a reinforcement learning agent to generate diverse programs based on symbolic instructions.

Researchers discuss connections and distinctions between DRITR and Kallus' work, focusing on policy evaluation and efficiency.

Machine learning identifies types of alterations in historical manuscripts.

Machine learning guides clinicians in predictive modeling using big data.

The paper proves an obstruction to immersed curves with specific curvature.

Paper introduces method to make neural networks symmetrical.

Proposes Neural SDE for better model robustness and generalization.

Graphs of neural networks are represented to preserve symmetry, improving performance across various tasks.

Convolutional Neural Processes improve data efficiency in neural processes.

Neural circuits integrate continuous dynamics efficiently.

GNPs use graph neural networks to predict target points with uncertainty quantification.

Investigates how neural network graph structure impacts predictive performance.

Neural Tangents simplifies infinite-width neural networks for research.

Novel framework explains generalization in deep neural networks.

This paper shows overparameterized deep neural networks are convex and learn useful features.

Augmented Neural ODEs improve stability and expressiveness of ODEs.

Neural networks can approximate functions uniformly across various measures.

Investigates neural codes and their embeddings, proving conjectures and introducing new code types.

Graph Metanetworks process diverse neural architectures efficiently.

Quadratic models explain neural network behavior during training.

Optimal rates for shallow ReLU networks in nonparametric regression.

New metric compares noisy neural trajectories using optimal transport.

Paper analyzes Q-learning with neural networks, proving a fast convergence rate.

Neural model guides PBE problem solving in programming.

The neural tangent kernel equivalence theorem fails in practice.