Request for open-sourcing the LLaVA w/DPO model weights

[pspdada]

I noticed that in your paper HSA-DPO, you trained an LLaVA w/DPO model as a comparison to demonstrate the effectiveness of your method. Although this model is not from your provided approach HSA-DPO, it has significant comparative value. Would you be able to open-source the weights of this model? It would be of great help to my research.

[Mr-Loevan]

I have uploaded the model to ModelScope. However, due to a file system failure, I cannot confirm whether it is the original version.

[pspdada]

Thanks!