TapTechNews May 16th news, according to the official Weibo of Xiaomi, today Xiaomi's large language model MiLM has officially been filed with the large model, and the related model is said to be gradually applied to Xiaomi's cars, phones, smart homes and other products, and in the future will be open to more users for experience.
Xiaomi's MiLM large model first appeared in the C-Eval and CMMLU large model evaluation list in August last year, ranking 10th on the C-Eval total list and 1st in the same order of magnitude at that time.
TapTechNews learned from the information on the GitHub project page that MiLM-6B is a large-scale pre-training language model developed by Xiaomi, with 64 billion parameters.
According to the information given by C-Eval, the MiLM-6B model has achieved high accuracy in various subjects, including metrology, physics, chemistry, biology and more in all 20 subjects of STEM (Science, Technology, Engineering and Mathematics education).
It is reported that the MiLM-6B model has shown relatively good accuracy in most liberal arts subjects, but there is still room for improvement in subjects involving abstract thinking such as law, mathematics, programming, probability theory, and discrete mathematics.
Among the 10 social science test subjects, except for education and geography, the model has achieved ideal accuracy in the other 8 subjects. In the 11 humanities subjects, MiLM-6B has performed well in history and law.