A marriage of formal methods and LLMs seeks to harness the strengths of both.

Engineers at the University of California San Diego have developed a new way to train artificial intelligence systems to ...

In “The Great Math War,” Jason Socrates Bardi takes on a battle for the soul of numbers that divided the experts of its day. By Jordan Ellenberg Jordan Ellenberg, a professor of mathematics at the ...

A few months before the 2025 International Mathematical Olympiad (IMO) in July, a three-person team at OpenAI made a long bet that they could use the competition’s brutally tough problems to train an ...

This repository contains the paper with examples demonstrating that, unlike current LLMs which generate deductive reasoning solutions through Chain-of-Thought prompting and heuristic pattern-mapping ...

French AI lab Mistral is getting into the reasoning AI model game. On Tuesday morning, Mistral announced Magistral, its first family of reasoning models. Like other reasoning models — e.g. OpenAI’s o3 ...

Americans think schools need to teach math skills to prepare students for future success—but they don’t always agree on which skills are the most crucial, and the subject stirs up complicated emotions ...

Google DeepMind has used chatbot models to come up with solutions to major problems in mathematics and computer science. The system, called AlphaEvolve, combines the creativity of a large language ...

Microsoft has introduced a new set of small language models called Phi-4-reasoning, Phi-4-reasoning-plus, and Phi-4-mini-reasoning, which are described as "marking a new era for efficient AI." These ...

What's the best way for children to learn arithmetic—memorizing number values and multiplication tables, or studying math at a deeper, conceptual level? Educators have long debated the merits of these ...

Despite great performance on Olympiad-level reasoning problems, frontier large language models can still struggle on high school math. We study the nature of language models’ (LM) reasoning by ...