Every month, several thousand academic papers are uploaded to the website arXiv.org, a website maintained by the Cornell University Library. Among them are the latest developments in machine learning, and other areas of research.
In December, 9,599 papers appeared on arXiv (pronounced “archive”), which works out to 309 papers per day on average. Among December’s postings: ProjectQ: An Open Source Software Framework for Quantum Computing, Brain-Swarm Interface (BSI): Controlling a Swarm of Robots with Brain and Eye Signals from an EEG Headset, and Unsupervised Video Segmentation via Spatio-Temporally Nonlocal Appearance Learning.
These papers show up before getting peer reviewed or published in prestigious journals. It is a veritable dumping ground for new, and recycled, ideas. Researchers at big companies like Facebook and Google post their latest on arXiv all the time, right alongside people working at universities around the world.
It’s not easy to keep up with arXiv, because of all the volume, and because the interface isn’t well suited for finding what’s relevant to you. There are no arXiv mobile apps — well, official ones, anyway.
So Amine Ben Khalifa, a research scientist at enterprise software company Workday, did the trendy thing and built an Alexa skill that lets you hear the titles and even abstracts of the 50 newest arXiv in the handful of research areas that are pertinent to him. So now he doesn’t have to set aside time every day to scan arXiv — he can just ask his Amazon Echo to tell him about stuff verbally. And if any papers sound interesting, he can review them in the Alexa app.
This is not the first attempt to simplify arXiv monitoring. Andrej Karpathy, a prominent figure in artificial intelligence (AI) who works at OpenAI, in 2015 introduced Arxiv Sanity Preserver, which not only shows a nice and neat presentation of machine learning papers on arXiv but also offers recommendations based on the papers that you “save.” In fact, Arxiv Sanity Preserver served as inspiration for the Alexa skill, which goes by the name arxivML, Khalifa wrote in a Medium post. There is also SciRate, which was once described as “Digg for the arXiv.”
arxivML is among the more than 7,000 skills that are now available for Alexa. Meanwhile Google, which fields the Google Home smart speaker that competes with the Echo, has been slowly rolling out “actions” for the Google Assistant that challenges Alexa.