Xinchen Zhang

> wondering the performance on sdxl ... > that would be great, thank you! We have extended BoxDiff to SDXL, please refer to [BoxDiff-XL](https://github.com/Cominclip/BoxDiff-XL). Note that the results are unstable...

> Hello, I'm interested in your perfect work. And I evaluated your method on the T2I-Compbench. The results are far from what you showed in the paper. I wonder if...

The link indicates that I need to have access permission. I have already requested permission from you using my email. If you are unable to grant access, you can send...

Thank you for your question. Due to the particularity of this benchmark, different random seeds have a significant impact on the test results for generating images. Based on the original...

> > Thank you for your question. Due to the particularity of this benchmark, different random seeds have a significant impact on the test results for generating images. Based on...

> Traceback (most recent call last): File "inference.py", line 334, in run(meta, args, starting_noise) File "inference.py", line 275, in run samples_fake = sampler.sample(S=steps, shape=shape, input=input, uc=uc, guidance_scale=config.guidance_scale, mask=inpainting_mask, x0=z0) File...

> Why does the script `scripts/train_refl_sdxl.sh` require images too (`train_refl.sh` doesnt)? I meet the same question, do you have solved it? i don't understand what the images refer to..

If you want to generate images only based on a text prompt, you should follow Option 1 and provide an api_key. This is because the layout is generated from GPT-4...

Hi, Thank you for your interest, and apologies for the delayed response. We’ve added a new file that allows you to directly use **OmniVerifier-7B** for verifying any generated images. You...

Thank you for your interest in this project! Omniverifier supports inference with VLLM, and we provide a [reference script](https://github.com/Cominclip/OmniVerifier/blob/main/inference_vllm.py) to help you get started. If you have any further questions,...