Series of example of accurate and inaccurate matching images to text
Language

Crawl the Web, Absorb the Bias: Language Models Absorb Biases from Web Training Data

The emerging generation of trillion-parameter models needs datasets of billions of examples, but the most readily available source of examples on that scale — the web — is polluted with bias and antisocial expressions. A new study examines the issue.
2 min read
GIF showing an orchestra playing music
Language

Roll Over, Beethoven: AI Completes Beethoven's 10th Symphony

Ludwig van Beethoven died before he completed what would have been his tenth and final symphony. A team of computer scientists and music scholars approximated the music that might have been.
2 min read
Animation showing gMLP, a simple architecture that performed some language and vision tasks as well as transformers
Language

Perceptrons Are All You Need: Google Brain's Multi-Layer Perceptron Rivals Transformers

The paper that introduced the transformer famously declared, “Attention is all you need.” To the contrary, new work shows you may not need transformer-style attention at all.What’s new: Hanxiao Liu and colleagues at Google
2 min read
Series of images showing some of the findings of the new study by researchers at Stanford’s Human AI Institute
Language

Weak Foundations Make Weak Models: Foundation AI Models Pass Flaws to Fine-Tuned Variants

A new study examines a major strain of recent research: huge models pretrained on immense quantities of uncurated, unlabeled data and then fine-tuned on a smaller, curated corpus.
2 min read
Graph showing Expire-span which enables attention to ignore tokens that aren’t useful to the task at hand
Language

Sharper Attention

Self-attention enables transformer networks to track relationships between distant tokens — such as text characters — in long sequences, but the computational resources required grow quadratically with input size.
2 min read
AI system monitoring urban traffic
Language

Meet the New Smart-Cities Champ

Chinese researchers for the first time swept a competition to develop AI systems that monitor urban traffic. Chinese universities and companies won first and second place place in all five categories of the 2021 AI City Challenge.
1 min read
Frozen Pretrained Transformer (FPT) explained
Language

Transformers: Smarter Than You Think

The transformer architecture has shown an uncanny ability to model not only language but also images and proteins. New research found that it can apply what it learns from the first domain to the others.
2 min read
Animated video showing a system to interpret electrical impulses from the brain as words
Language

Listening to the Brain

Neural networks translated a paralyzed man’s brainwaves into conversational phrases. Researchers trained a system to interpret electrical impulses from the brain of a man who had lost the ability to speak 15 years ago, and displayed them as words on a video screen.
1 min read
Screen captures of a job interviews automation system
Language

Danke for the Interview

An independent test found flaws in AI systems designed to evaluate job applicants. MyInterview and Curious Thing, which automate job interviews, gave a candidate who spoke only in German high marks on English proficiency, according to MIT Technology Review.
2 min read
Series of AI generated imagery
Language

CLIP Art

Creative engineers are combining deep learning systems to produce a groundswell of generated imagery. Researchers, hackers, and artists are producing new works by pairing CLIP, a pretrained image classifier, with a generative adversarial network (GAN).
1 min read
Image showing how object detectors work
Language

I Know It When I See It

Object detectors typically detect only items that were labeled in their training data. A new method liberates them to locate and recognize a much wider variety of objects.
2 min read
New approach to automated real-estate assessment
Language

Home Sweet AI-Appraised Home

Real estate websites helped turn automated real-estate assessment into a classic AI problem. The latest approach by a leader in the field gets a boost from deep learning.
2 min read
Architecture of vision-language tasks
Language

One Model for Vision-Language

Researchers have proposed task-agnostic architectures for image classification tasks and language tasks. New work proposes a single architecture for vision-language tasks.
2 min read
Libel-detection system working
Language

Double Check for Defamation

A libel-detection system could help news outlets and social media companies stay out of legal hot water. CaliberAI, an Irish startup, scans text for statements that could be considered defamatory, Wired reported.
2 min read
Walking through a narrow hallway in a library
Language

Bias By the Book

Researchers found serious flaws in an influential language dataset, highlighting the need for better documentation of data used in machine learning.
1 min read

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox