Récemment, Intelligence artificielle (AI) has reached a historic milestone in one of the world’s toughest math contests, the International Mathematical Olympiad (IMO). Google DeepMind’s Gemini Deep ...

Abstract: Multimodal Large Language Models (MLLMs) have shown promising capabilities in mathematical reasoning within visual contexts across various datasets. However, most existing multimodal math ...

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 ...

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 ...

There’s a curious contradiction at the heart of today’s most capable AI models that purport to “reason”: They can solve routine math problems with accuracy, yet when faced with formulating deeper ...

The Nature Index 2025 Research Leaders — previously known as Annual Tables — reveal the leading institutions and countries/territories in the natural and health sciences, according to their output in ...

Inductive reasoning is a critical skill that enables individuals to make sound decisions by drawing general conclusions from specific observations. Whether you’re working on a high-stakes business ...

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 ...

The big picture: Benchmarking AI remains a thorny issue, with companies often accused of cherry-picking flattering results while burying less favorable ones. Instead of fixating on math and logic ...

Shortly after OpenAI released o1, its first “reasoning” AI model, people began noting a curious phenomenon. The model would sometimes begin “thinking” in Chinese, Persian, or some other language — ...