Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizens for paving the road, the West window, the rain and snow, and the delivery of clues from Daniel Wu in South China! CTOnews.com August 11 news, Xiaomi company did not "official Xuan" directly joined the AI big model track, but Xiaomi AI big model MiLM-6B has quietly appeared in the C-Eval, CMMLU big model evaluation list, so far, Xiaomi big model in the C-Eval total list ranked 10th, the same parameter order of magnitude ranked first.

According to the information given on the GitHub project page, CTOnews.com knows that MiLM-6B is a large-scale pre-training language model developed by Xiaomi Company with a parameter scale of 6.4 billion.

The ▲ map is from the GitHub page of MiLM-6B, and according to the information given by C-Eval, the MiLM-6B model has achieved high accuracy in all 20 subjects of STEM (science, technology, engineering and mathematics education), such as metrology, physics, chemistry, biology and so on.

It is learned from ▲ source C-Eval that the MiLM-6B model has shown relatively good accuracy in most liberal arts subjects, but there is still some room for improvement in "law, mathematics, programming, probability theory, discrete mathematics" and other subjects involving "abstract thinking."

Among the 10 "social science" test subjects, the model has achieved ideal accuracy in 8 other subjects except "Pedagogy and Geography". Among the 11 subjects in the humanities, MiLM-6B has a good accuracy performance on the basis of History and Law.

MiLM-6B, a large AI model, has only been exposed for the first time now, but it is already highly completed, and it may be used in virtual assistant services such as Xiaomi's Xiao Ai in the future.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.