Jordan Ellenberg
@JSEllenberg
Math professor at Wisconsin. Number enthusiast. Author of HOW NOT TO BE WRONG and SHAPE (May 2021)
ID:1087183776
http://jordanellenberg.com 13-01-2013 20:53:18
30,8K Tweets
28,1K Followers
907 Following
Follow People
New paper on label-efficient supervised finetuning of LLMs.
We address the expensive prompt annotation cost by humans/proprietary LLMs, saving as much as 50% on FLAN V2.
Paper: arxiv.org/abs/2401.06692
Work led by: Jifan Zhang Yifang Chen Gantavya Bhatt Arnav Das
1/