Apple releases details about new 'MM1' AI model.

Apple researchers have developed a new way to train large language models (LLMs) that seamlessly integrate text and visual information.

The company's findings are detailed in a research paper titled “MM1: Methods, analysis, and insights from Multimodal LLM pre-training“, showcases a new approach to creating smarter, more flexible AI systems. By using a diverse dataset that includes image caption pairs, image-nested text documents, and text-only data, Apple claims the MM1 model sets a new standard in AI's ability to perform… Tasks such as image captioning, visual question answering, and natural language inference with a high level of accuracy.

Apple's research focuses on combining different types of training data and model architectures, enabling AI to understand and generate language based on a combination of visual and linguistic cues. This ability is vital for tasks that require a precise understanding of the world, such as interpreting complex images or answering questions involving visual elements.

The paper also highlights the exceptional context learning capabilities of the MM1 model, particularly at the model's largest parameter configuration of 30 billion. This release apparently demonstrates impressive capabilities for multi-step inference on multiple images using a few-shot “chain of thought,” a technique that allows AI to perform complex, open-ended problem solving based on minimal examples.

This research appears as part of Apple's broader initiative to enhance its capabilities in the field of artificial intelligence amid increasing competition. Earlier today, BloombergApple is in discussions with Google to license Gemini's large-scale generative models from Google to power new features coming to the iPhone as part of iOS 18, Apple's Mark Gurman reported.