user
Hopehope赋予希望
03-18

$XIAOMI-W(01810)$  


Xiaomi released an SOTA audio reasoning model which has been developed by applying DeepSeek-R1's GRPO RL algo. Achieved 64.% on MMAU benchmark after just 1 wk.

Xiaomi made break thru applying GRPO to Qwen2-Audio-7B model using 38000 training samples from Tsinghua's AVQA dataset.

R1 has made a profound impact in China's entire AI ecosystem.


source: tphuang

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Comments

Leave a comment
1