PKU-YuanGroup / Video-LLaVA

【EMNLP 2024🔥】Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
https://arxiv.org/pdf/2311.10122.pdf
Apache License 2.0
3.03k stars 220 forks source link

Video-LLava Upgradation #164

Open Tortoise17 opened 6 months ago

Tortoise17 commented 6 months ago

Dear Friends, I have to ask that is there any plan to retrain it with LLAMA-3 based model? to improve its working performance. or if there is already any available? If you could kindly guide and help.

hongkunhao commented 5 months ago

Same question. I tried to retrain it with LLAMA-3, but there are library mismatch questions. Could anyone give some guide? Thanks!