Generative Model

Minimal quickstart fork of the LLaDA repository. Watch text diffusion with a single line of code (& flag). - TheFloatingString/LLaDA-minimal-quickstart

MIT CSAIL's "MDGen" system takes a frame of a 3D molecule and simulates what will happen next, connects separate stills, or fills in missing frames. The generative model "presses play" on molecules, potentially helping chemists design new molecules and drug prototypes.

You already know that agents and small language models are the next big things. Here are five other hot trends you should watch out for this year.

Deloitte survey shows business and IT leaders are optimistic, but academic researchers warn of AI training time bomb

Following a report from The Wall Street Journal that claims OpenAI has been sitting on a tool that can spot essays written by ChatGPT with a high degree of...

This paper describes a discrete state-space model -- and accompanying methods -- for generative modelling. This model generalises partially observed Markov decision processes to include paths as latent variables, rendering it suitable for active inference and learning in a dynamic setting. Specifically, we consider deep or hierarchical forms using the renormalisation group. The ensuing renormalising generative…

Instead, Meta's Yann LeCun says, work on next-gen AI systems that lift the limitations of LLMs. But what are those next-gen systems?

Google has expanded its availability of the Gemini 1.5 Pro LLM, making its audio processing capability accessible to more enterprise users.

Active inference is a unifying theory for perception and action resting upon the idea that the brain maintains an internal model of the world by minimizing free energy. From a behavioral perspective, active inference agents can be seen as self-evidencing beings that act to fulfill their optimistic predictions, namely preferred outcomes or goals. In contrast, reinforcement learning requires human-designed rewards to…

How do ChatGPT and the other contemporary AIs look when compared with this understanding of human brains and human minds?

MIT researchers developed a way to quickly calculate the transition state structure of a chemical reaction, using machine-learning models.

A new AI tool that predicts viral mutations could guide treatments for COVID-19 — and the next pandemic. The system, named EVEscape, was developed at Harvard Medical School and ...

Neural network AI programs that are reduced by 75% of their functionality can achieve the same power as programs twice their size.

Great news everyone! Amazon's Alexa is getting a slew of new AI-enhanced abilities!

The spotlight has been on language and images for Generative AI, but there's been a lot of recent progress in the audio domain. Learn everything you need to know about generative audio models in this article.

Here's a puzzle: We've seen a lot of papers claiming you can use one language model to generate useful training data for another language model. But…by the data processing inequality, we shouldn't expect to be able to create new information that wasn't in the first model's training set.

Adversarial watermarking has been proposed many times as a preventative measure against deepfaking, but usually with the intention to entirely disrupt the generative process. But a new method proposes that it could be used to give individuals more control over the use of their identities in such systems.

Machine learning models can help to solve several real-world problems faster and more efficiently. One among these problems involves synthesizing speech for both animated characters and human speakers based on the movements of their lips.

The BAIR Blog

A Perspective from the Frontiers in AI Art