Nvidia's Nemotron-Cascade 2 wins math and coding gold medals with 3B active parameters — and its post-training recipe is now open-source

Nvidia's Nemotron-Cascade 2 is a 30B MoE model that activates only 3B parameters at inference time, yet achieved gold ...
Fast Company

Are LTMs the next LLMs? This new type of AI can do what large-language models can’t

Large-language models (LLMs) have taken the world by storm, but they’re only one type of underlying AI model. An under-the-radar company, Fundamental, is set to bring a new type of enterprise AI model ...
Unite.AI

Verbosity Decreases Accuracy in Large Language Models

New research finds that forcing Large Language Models to give shorter answers notably improves the accuracy and quality of ...
The American Journal of Managed Care

Health Equity in the Era of Large Language Models

This article presents challenges and solutions regarding health care–focused large language models (LLMs) and summarizes key recommendations from major regulatory and governance bodies for LLM ...

Three ways AI is learning to understand the physical world

Large language models lack grounding in physical causality — a gap world models are designed to fill. Here's how three distinct architectural approaches (JEPA, Gaussian splats, and end-to-end ...
MIT Technology Review

Meet the new biologists treating LLMs like aliens

How large is a large language model? Think about it this way. In the center of San Francisco there’s a hill called Twin Peaks from which you can view nearly the entire city. Picture all of it—every ...