Facebook’s leadership has thwarted changes in its algorithms aimed at making the site less polarizing, according to the Wall Street Journal. The social network’s own researchers determined that its AI software promotes divisive content.

Facebook recently rolled out its entry in the World’s Biggest Chatbot sweepstakes. In keeping with the company’s social-networking dominance, the bot is designed to excel at chitchat on any subject.

Facebook’s AI can’t spot Covid-19 disinformation on its own. But with human help, it can slow the spread. Facebook uses a combination of humans and neural nets to crack down on messages that make false claims about Covid-19, which may have deadly consequences.

Batch normalization is a technique that normalizes layer outputs to accelerate neural network training. But new research shows that it has other effects that may be more important.

Which dataset was used to train a given model? A new method makes it possible to see traces of the training corpus in a model’s output.

Geoffrey Hinton, Yoshua Bengio, and Yann LeCun presented their latest thinking about deep learning’s limitations and how to overcome them.

School teachers may not like to hear this, but sometimes you get the best answer by peeking at your neighbor’s paper. A new language model framework peeks at the training data for context when making a prediction.

In tasks that involve generating natural language, neural networks often map an input sequence of words to an output sequence of words. Facebook researchers used a similar technique on sequences of mathematical symbols, training a model to map math problems to math solutions.

Facebook announced a ban on deepfake videos, on the heels of a crackdown on counterfeit profiles that used AI-generated faces. Facebook declared this week that it will remove deepfake videos it deems deliberately misleading.

How is it that many people learn to drive a car fairly safely in 20 hours of practice, while current imitation learning algorithms take hundreds of thousands of hours, and reinforcement learning algorithms take millions of hours? Clearly we’re missing something big.

Earlier language models powered by Word2Vec and GloVe embeddings yielded confused chatbots, grammar tools with middle-school reading comprehension, and not-half-bad translations. The latest generation is so good, some people consider it dangerous.

Society awakened to the delight, threat, and sheer weirdness of realistic images and other media dreamed up by computers.

As language models show increasing power, a parallel trend has received less notice: The vogue for naming models after characters in the children’s TV show Sesame Street.

Models that summarize documents and answer questions work pretty well with limited source material, but they can slip into incoherence when they draw from a sizeable corpus. Recent work addresses this problem.

A word-embedding model typically learns vector representations from a large, general-purpose corpus like Google News. But to make the resulting vectors useful in a specialized domain, they must be fine-tuned on a smaller, domain-specific dataset. Researchers offer a more accurate method.