HN2new | past | comments | ask | show | jobs | submitlogin

I wonder whether LoRAs could be useful for U-Net training. Especially thinking of CNN-based U-Net models with pre-trained encoders (but randomly initialized decoders). At least, it seems possible that normal weight updates on the decoder and LoRA training on the encoder could improve efficiency.


Diffusion unet has an "extended" version nowadays that applies to the resnet part as well as the cross-attention: https://github.com/cloneofsimo/lora




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: