Researchers have developed a new method for training AI models to improve mathematical reasoning. Instead of just rewarding the correct final answer, they've used 'process supervision' to encourage each step of reasoning. This approach has achieved state-of-the-art performance and also helps align the model's thought process with human standards. --- Why it matters: This matters because it shows a new way to train AI models to think more like humans, which is crucial for tasks that require complex mathematical reasoning such as scientific research or financial analysis. Source: https://openai.com/index/improving-mathematical-reasoning-with-process-supervision