Paper?

#6
by Ano-Nimus - opened

First, thank you for open sourcing this project.

I find it really interesting that you have come up with a way to produce a good LoRA in such a short amount of time and with a small dataset. I would like to read more on exactly how this was done.

Will there be a paper release?

DiffSynth-Studio org

Thank you for your interest!
We've detailed our complete approach in our blog post here: Qwen-Image-i2L: Training Strategies for Image-to-LoRA Generation.

Sign up or log in to comment