Classification

150 Posts

Language Models Defy Logic: Large NLP models struggle with logical reasoning.
Classification

Language Models Defy Logic: Large NLP models struggle with logical reasoning.

Who would disagree that, if all people are mortal and Socrates is a person, Socrates must be mortal? GPT-3, for one. Recent work shows that bigger language models are not necessarily better when it comes to logical reasoning.
High-level overview of the STEGO architecture at train and prediction steps
Classification

Segmented Images, No Labeled Data: Improved unsupervised learning for semantic segmentation

Training a model to separate the objects in a picture typically requires labeled images for best results. Recent work upped the ante for training without labels.
Map showing areas likely to have been damaged by Hurricane Leo
Classification

Wreckage Recognition: AI System Spots Hurricane Ian Damage

A machine learning model identified areas likely to have been damaged by Hurricane Leo as it swept through the southern United States.
Capture of Honorlock, an AI-powered software designed to catch students who cheat on academic examinations
Classification

Court Blocks AI-Assisted Proctoring: AI Cheating Software Loses in Court

A U.S. court ruled against an implementation of AI-powered software designed to catch students who cheat on academic examinations.
Object-Detection Transformers Simplified: New Research Improves Object Detection With Vision Transformers
Classification

Object-Detection Transformers Simplified: New Research Improves Object Detection With Vision Transformers

ViTDet, a new system from Facebook, adds an object detector to a plain pretrained transformer.
Animated flowcharts show how the ProtCNN AI model classifies proteins.
Classification

Protein Families Deciphered: Machine Learning Categorizes Proteins Based on Their Functions

Convolutional neural networks separate proteins into functional families without considering their shapes.
An animation shows an AI-powered system called From Numbers to Names, which identifies Holocaust victims in photographs.
Classification

Identifying Faces of History: From Numbers to Names Uses AI to Identify Holocaust Victims

From Numbers to Names matches individuals to faces in publicly available images related to the genocide of European Jews between 1941 and 1945.
Bees pollinating flowers
Classification

Protection for Pollinators: AI Could Help Create Pesticides That Don’t Kill Bees

A machine learning method could help chemists formulate pesticides that target harmful insects but leave bees alone. Researchers at Oregon State University developed models that classify whether or not a chemical is fatally toxic to bees.
A flowchart shows how a jury learning method reduces annotator bias in machine learning models.
Classification

Choose the Right Annotators: Jury-Learning Helps Remove Bias from NLP Models

A new machine learning method attempts to account for biases that may be held by certain subsets of labelers.
Flowcharts show how a new contrastive learning approach uses metadata to improve AI image classifiers
Classification

Learning From Metadata: Descriptive Text Improves Performance for AI Image Classification Systems

Images in the wild may not come with labels, but they often include metadata. A new training method takes advantage of this information to improve contrastive learning.
A football player performs drills on the pitch while a computer vision system tracks and grades his movements
Classification

On the Ball: An AI-Powered App Lets Amateur Footballers Try Out for the Pros

AiSCOUT uses computer vision to grade amateur footballers and recommends those who score highest to representatives of professional teams.
A Cadillac SUV drives through one of UVeye's Atlas arches
Classification

Auto Diagnosis: AI-Powered Inspections Arrive at Dealers for GM and Volvo

A drive-through system from UVeye automatically inspects vehicles for dents, leaks, and low tire pressure.
Masked Auto-Encoder (MAE) explanation
Classification

Who Was That Masked Input? Pretraining Method Improves Computer Vision Performance

Researchers have shown that it’s possible to train a computer vision model effectively on around 66 percent of the pixels in each training image. New work used 25 percent, saving computation and boosting performance to boot.
Satalite Map image
Classification

Tracking Changes on Earth’s Surface: AI Tracks Land Change from Satellite Imagery

Computer vision systems are scanning satellite photos to track construction on the Earth’s surface — an exercise in behavior recognition on a global scale.
Smart Phone app called Dieta
Classification

Deep Doo-Doo: AI App Diagnoses Poop Better Than People

People who suffer from gastrointestinal conditions such as irritable bowel syndrome are number two when it comes to describing the characteristics of their own poop.

Subscribe to The Batch

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