The LLM craze initiated by OpenAI continues to rise rapidly with Apple and Google entering the artificial intelligence market. Artificial intelligence researchers working at Apple published an article about model architectures and performance improvements under the title “MM1: Methods, Analysis and Insights from Multimodal LLM Pre-Training”.
Apple published an article about performance improvements in the MM1 model
The paper, published this week on arxiv.org by Apple employees, includes information about how carefully combining different types of training data and model architectures can lead to Emphasizing the importance of scaling visual components in the article, the team underlined that the choice of image encoder and the resolution of the input images have a great impact on model performance.
“For large-scale multimodal pre-training, we show that using a careful mix of caption, spaced image text, and text-only data is crucial to achieving state-of-the-art multi-shot results across multiple benchmarks,” the published paper states. ” statements were included.
Is artificial intelligence becoming a doctor? Diagnosis of ear infection
Is artificial intelligence now trying to become a doctor? Researchers have developed an ear infection diagnostic app that uses artificial intelligence.
The team found that the image encoder had a significant impact along with the image resolution and number of image tokens, while the image language connector design was of relatively negligible importance. The largest (30 billion parameters) MM1 model demonstrated strong in-context learning capabilities, allowing it to perform multi-step reasoning on multiple input images using several steps of “chain of thought” guidance.
The outcome of the research indicates the potential of large multimodal models to tackle complex, open-ended problems that require deep-rooted language understanding and production. The MM1 research comes at a time when Apple is accelerating its investments in artificial intelligence to catch up with rivals such as Google, Microsoft and Amazon, which are advancing to integrate productive artificial intelligence capabilities into their products.
The company is on track to spend $1 billion a year on AI development, according to a recent report from Bloomberg. Sources say Apple is working on a large language model framework called “Ajax” as well as a chatbot known internally as “Apple GPT.”
It is stated that the company plans to integrate these technologies into Siri, Messages, Apple Music and other applications and services. For example, AI can be used to automatically create personalized playlists, assist developers with writing code, or engage in open-ended conversation and task completion.