Hey CV friends, we’re hosting our weekly live discussion

ArXiv Dives

. This week we’ll cover

How DeepSeek-R1 used GRPO for Reinforcement Learning

, building upon last week’s DeepSeek paper review (if you want to see that video on How R1 and GRPO Work - deep technical dive into DeepSeek’s Models, check that on ). We’re live today at 10amPT… grab your coffee and come join the convo! https://lu.ma/arxivdive-36