Hao Jiang

Train with w-plus pSp encoder.

2

I'm wondering if this repository can be trained with w-plus pSp encoder since original pSp is designed to predict wplus latent code from input image. I have read the DualStyleGAN...

Why the mid result of convolution from C*h1*w1 is C*1*1 instead of d*1*1?

1

Hello, recently i have read your paper A2U and been amazed, but i have a problem while figuring out the process of **upsampling kernel generation**. In Eq.3, we can learn...

It seems not proper to use `skimage.exposure.rescale_intensity` to adjust exposure of final output in test.py, which causes dark enhancement results if there are over-exposured pixels, it's better to clip intensity...

The usage of past_key_values in AutoCompressorMixin

1

Hello authors. There is no doubt that AutoCompressors is an excellent work. I have carefully read and learned the mainly designed codes in `auto_compressor.py` and noticed that the class `AutoCompressorMixin`...

Does Auto-Compressors support newer base model like LLama-3 or Qwen-2.5?

2

I have tried to train with newer base model LLama-3-3B, and upgrade dependent libraries like transformers, accelerate and so on. But still fails to train.

Excellent work for compressing content length. This work is very similar to AutoCompressors (which is mentioned in this paper), one difference is that AutoCompressors encode/decode compression by LoRA-LLM while ICAE...