张志鸿(zhang zhihong)

https://www.ustc.edu.cn/

University of Science and Technology of China Hefei, China Student of School of Information Science and Technology, University of Science and Technology of China

Results 2 comments of


                                            张志鸿(zhang zhihong)

dpo_ov7b.sh训练问题

CUDA out of memory +1

Request of code or guideline to generate preference data using llava critic. (==reproduce llava-onevision-ov-chat)

+1, @tyxiong23 , The llava-rlhf dataset is multi-turn form, how to choose question-image pairs to sample responses.