Networks

Dynamic Routing Between Capsules

Sara Sabour, Nicholas Frosst, Geoffrey E Hinton

A capsule is a group of neurons whose activity vector represents the instantiation parameters of a specific type of entity such as an object or an object part.

The Curse of Recursion: Training on Generated Data Makes Models Forget

Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao +3

Stable Diffusion revolutionised image creation from descriptive text.

How to factor 2048 bit RSA integers in 8 hours using 20 million noisy qubits

Craig Gidney, Martin Ekerå

We significantly reduce the cost of factoring integers and computing discrete logarithms in finite fields on a quantum computer by combining techniques from Shor 1994, Griffiths-Niu 1996, Zalka 2006, Fowler 2012, Ekerå-Håstad 2017, Ekerå 2017, Ekerå 2018, Gidney-Fowler 2019, Gidney 2019.

Responses to Critiques on Machine Learning of Criminality Perceptions (Addendum of arXiv:1611.04135)

Xiaolin Wu, Xi Zhang

In November 2016 we submitted to arXiv our paper "Automated Inference on Criminality Using Face Images".

Learning Transferable Visual Models From Natural Language Supervision

Alec Radford, Jong Wook Kim, Chris Hallacy +9

State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories.

Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity

William Fedus, Barret Zoph, Noam Shazeer

In deep learning, models typically reuse the same parameters for all inputs.

Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking

Eric Zelikman, Georges Harik, Yijia Shao +3

When writing and talking, people sometimes pause to think.

Internet encyclopaedias go head to head

Jim Giles

Building high-level features using large scale unsupervised learning

Quoc V. Le, Marc'Aurelio Ranzato, Rajat Monga +5

We consider the problem of building high-level, class-specific feature detectors from only unlabeled data.

Don't Skype & Type! Acoustic Eavesdropping in Voice-Over-IP

Alberto Compagno, Mauro Conti, Daniele Lain +1

Acoustic emanations of computer keyboards represent a serious privacy issue.

Ending the Anomaly: Achieving Low Latency and Airtime Fairness in WiFi

Toke Høiland-Jørgensen, Michał Kazior, Dave Täht +2

With more devices connected, delays and jitter at the WiFi hop become more prevalent, and correct functioning during network congestion becomes more important.

Scaling Data-Constrained Language Models

Niklas Muennighoff, Alexander M. Rush, Boaz Barak +6

The current trend of scaling language models involves increasing both parameter count and training dataset size.

QUIC is not Quick Enough over Fast Internet

Xumiao Zhang, Shuowei Jin, Yi He +4

QUIC is expected to be a game-changer in improving web application performance.

Banishing LLM Hallucinations Requires Rethinking Generalization

Johnny Li, Saksham Consul, Eda Zhou +9

Despite their powerful chat, coding, and reasoning abilities, Large Language Models (LLMs) frequently hallucinate.

Three Attacks on Proof-of-Stake Ethereum

Caspar Schwarz-Schilling, Joachim Neu, Barnabé Monnot +3

Recently, two attacks were presented against Proof-of-Stake (PoS) Ethereum: one where short-range reorganizations of the underlying consensus chain are used to increase individual validators' profits and delay consensus decisions, and one where adversarial network delay is leveraged to stall consensus decisions indefinitely.

A Benchmark for Learning to Translate a New Language from One Grammar Book

Garrett Tanzer, Mirac Suzgun, Eline Visser +2

Large language models (LLMs) can perform impressive feats with in-context learning or lightweight finetuning.

Large-scale online deanonymization with LLMs

Simon Lermen, Daniel Paleka, Joshua Swanson +3

We show that large language models can be used to perform at-scale deanonymization.

Exposure to the Russian Internet Research Agency foreign influence campaign on Twitter in the 2016 US election and its relationship to attitudes and voting behavior

Gregory Eady, Tom Paskhalis, Jan Zilinsky +3

Abstract There is widespread concern that foreign actors are using social media to interfere in elections worldwide.

The Spy in the Sandbox -- Practical Cache Attacks in Javascript

Yossef Oren, Vasileios P. Kemerlis, Simha Sethumadhavan +1

We present the first micro-architectural side-channel attack which runs entirely in the browser.

Fansmitter: Acoustic Data Exfiltration from (Speakerless) Air-Gapped Computers

Mordechai Guri, Yosef Solewicz, Andrey Daidakulov +1

Because computers may contain or interact with sensitive information, they are often air-gapped and in this way kept isolated and disconnected from the Internet.

BrainNet: A Multi-Person Brain-to-Brain Interface for Direct Collaboration Between Brains

Linxing Preston Jiang, Andrea Stocco, Darby M. Losey +3

We present BrainNet which, to our knowledge, is the first multi-person non-invasive direct brain-to-brain interface for collaborative problem solving.

Comparing the Effects of DNS, DoT, and DoH on Web Performance

Austin Hounsel, Kevin Borgolte, Paul Schmitt +2

Nearly every service on the Internet relies on the Domain Name System (DNS), which translates a human-readable name to an IP address before two endpoints can communicate.

Extracting Training Data from Large Language Models

Nicholas Carlini, Florian Tramer, Eric Wallace +9

It has become common to publish large (billion parameter) language models that have been trained on private datasets.

Genie: Generative Interactive Environments

Jake Bruce, Michael Dennis, Ashley Edwards +22

We introduce Genie, the first generative interactive environment trained in an unsupervised manner from unlabelled Internet videos.

Ethernet

Robert M. Metcalfe, David R. Boggs

Ethernet is a branching broadcast communication system for carrying digital data packets among locally distributed computing stations.