On Inference Stability for Diffusion Models

January 8, 2024

Denoising Probabilistic Models (DPMs) represent an emerging domain of generative models that excel in generating diverse and high-quality images. However, most current training methods for DPMs often neglect the correlation between timesteps, limiting the model’s performance in generating images effectively. Notably, we theoretically point out that this issue can be caused by the cumulative estimation gap between the predicted and the actual trajectory. To minimize that gap, we propose a novel sequence-aware loss that aims to reduce the estimation gap to enhance the sampling quality. Furthermore, we theoretically show that our proposed loss function is a tighter upper bound of the estimation loss in comparison with the conventional loss in DPMs. Experimental results on several benchmark datasets including CIFAR10, CelebA, and CelebA-HQ consistently show a remarkable improvement of our proposed method regarding the image generalization quality measured by FID and Inception Score compared to several DPM baselines. Our code and pre-trained checkpoints are available at


< 1 minute

Viet Nguyen*, Giang Vu*, Tung Nguyen Thanh, Khoat Than, Toan Tran

Share Article

Related publications

ML ICLR Top Tier
February 19, 2024

Nguyen Hung-Quang, Yingjie Lao, Tung Pham, Kok-Seng Wong, Khoa D Doan

January 8, 2024

Tran Huynh Ngoc, Dang Minh Nguyen, Tung Pham, Anh Tran

ML AAAI Top Tier
January 8, 2024

Viet Nguyen*, Giang Vu*, Tung Nguyen Thanh, Khoat Than, Toan Tran

ML NeurIPS Top Tier
October 4, 2023

Van-Anh Nguyen, Trung Le, Anh Tuan Bui, Thanh-Toan Do, Dinh Phung