Request for Additional Code, Random Seed, or Specific Style Prompt Information to Reproduce Qualitative Results

[curryjung]

I am currently engaged in research that builds upon the concepts presented in your paper. In order to validate our findings and compare them with your results, we are attempting to reproduce the qualitative results from your paper. However, we have encountered difficulties due to the unavailability of certain codes, random seeds, or specific style prompt information.

Therefore, I kindly request your assistance in providing the following:

1. Any additional code or scripts that were used to generate the qualitative results presented in your paper.
2. Information about the random seed(s) or saved initial Gaussian noise used in your experiments.
3. Any specific style prompt information that may be relevant to the qualitative results.

Having access to this information would greatly facilitate our efforts to reproduce and build upon your results, ensuring the validity and comparability of our research. Rest assured that any provided code or data will be used solely for the purpose of academic research.

Thank you very much for your attention to this matter. I look forward to your response and hope for the opportunity to discuss this further.

[amirhertz]

The list of prompts we used for evaluation is provided in the appendix of our paper (last two pages).
We generated the 4 "style aligned" images for each prompt using DDIM sampler with 50 steps.
The sampled noise for each image can be found here.

[david-alvarez-rosa]

Oh, boi.