LLaVA-NeXT
LLaVA-NeXT copied to clipboard
When will mm_use_im_start_end be implemented in pre-training?
Hi,
Why is this functionality not implemented?
https://github.com/LLaVA-VL/LLaVA-NeXT/blob/1654303050a429f5c0dbb97033bf24a2263531cf/llava/model/llava_arch.py#L233-L235
What is needed to implement this?
Best, Orr