Bidirectional Noise Injection: Enhancing Diffusion Models via Coordinated Input-Output Perturbation

Diffusion models have demonstrated remarkable success in image generation, yet a persistent challenge remains: the bias between model predictions and the target distribution. In this paper, we propose a Bidirectional Noise Injection framework for enhancing diffusion models, implemented via Coordinated Input-Output Perturbation (CIOP). Our approach mitigates this bias by randomly applying synchronized noise injection to both the model inputs and the prediction targets during the training stage. This stochastic, synchronized noise injected acts as a smoothing mechanism that effectively reduces the 2-Wasserstein distance between the predicted and target distributions, as substantiated by our theoretical analysis based on optimal transport theory. Extensive experiments on multiple benchmark datasets and various generative tasks demonstrate that our method improves generation quality and training efficiency without incurring additional computational cost. Furthermore, the design of CIOP enables seamless integration with existing diffusion model improvements and advanced frameworks, thereby broadening its applicability. These results highlight the potential of Bidirectional Noise Injection via CIOP to alleviate bias in diffusion-based generative models across a wide range of settings.

Authors

• Tianyi Zheng vivo Mobile Communication Co., Ltd
• Jiayang Gao Shanghai Jiao Tong University
• Peng-tao Jiang vivo Mobile Communication Co., Ltd
• Fengxiang Yang Department of Artificial Intelligence, Xiamen University, China Department of Information Engineering and Computer Science, University of Trento, Italy
• Ben Wan Shanghai Jiao Tong University
• Hao Zhang vivo Mobile Communication Co., Ltd
• Jinwei Chen vivo Mobile Communication Co., Ltd
• Jia Wang
• Bo Li Department of Computing, The Hong Kong Polytechnic University

Keywords

No keywords are indexed for this paper.