Representative Guidance: Diffusion Model Sampling with Coherence

Authors: Anh-Dung Dinh, Daochang Liu, Chang Xu

ICLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments are conducted to evaluate on Image Net Deng et al. (2009) dataset with two resolutions 64x64 and 256x256 with 50000 generated samples. We first verify our claims that our proposed Rep G helped to improve the details and fix the faulty information in the images qualitatively in section5.1 and quantitatively in section5.2. After that, we will compare quantitatively with other state-of-the-art methods such as Big GAN Brock et al. (2018), ADM Dhariwal & Nichol (2021), Px P Dinh et al. (2023a), Pro G Dinh et al. (2023b), EDS Zheng et al. (2022), IDDPM Nichol & Dhariwal (2021), VAQ-VAE-2 Razavi et al. (2019) and Classifier-free guidance (CLSFree) Ho & Salimans (2022).
Researcher Affiliation Academia Anh-Dung Dinh School of Computer Science The University of Sydney EMAIL Daochang Liu School of Physics, Mathematics and Computing The University of Western Australia EMAIL Chang Xu School of Computer Science The University of Sydney EMAIL
Pseudocode Yes Algorithm 1 DDPM denoising process with representative guidance
Open Source Code Yes source code: https://github.com/dungdinhanh/rep-guidance.
Open Datasets Yes Experiments are conducted to evaluate on Image Net Deng et al. (2009) dataset with two resolutions 64x64 and 256x256 with 50000 generated samples.
Dataset Splits Yes Experiments are conducted to evaluate on Image Net Deng et al. (2009) dataset with two resolutions 64x64 and 256x256 with 50000 generated samples.
Hardware Specification Yes All the experiments in this paper are conducted on A100 GPUs 40GB.
Software Dependencies No The paper mentions self-supervised models like Mo Co-v2, Sim Siam, and Moco-v3 by their scientific papers but does not specify any general software dependencies (e.g., Python, PyTorch, CUDA) with version numbers.
Experiment Setup Yes Table 6: All hyperparameters for producing the results are shown in this table. Model Datasets K H γ IDDPM + Rep G Image Net64x64 5 1 10.0 ADM + Rep G Image Net64x64 5 1 10.0 ADM-CLSFree + Rep G Image Net64x64 5 1 8.0 ADM + Rep G Image Net256x256 10 2 20.0 ADM-CLSFree + Rep G Image Net256x256 10 2 20.0 Di T-CLSFree + Rep G Image Net256x256 10 2 15.0