Bias mitigation in graph diffusion models
This addresses bias issues in graph diffusion models for improved generation quality, representing a strong incremental improvement.
The paper tackles bias problems in graph diffusion models, specifically reverse-starting and exposure biases, by proposing a comprehensive approach with a new Langevin sampling algorithm and score correction mechanism, achieving state-of-the-art results across multiple models, datasets, and tasks.
Most existing graph diffusion models have significant bias problems. We observe that the forward diffusion's maximum perturbation distribution in most models deviates from the standard Gaussian distribution, while reverse sampling consistently starts from a standard Gaussian distribution, which results in a reverse-starting bias. Together with the inherent exposure bias of diffusion models, this results in degraded generation quality. This paper proposes a comprehensive approach to mitigate both biases. To mitigate reverse-starting bias, we employ a newly designed Langevin sampling algorithm to align with the forward maximum perturbation distribution, establishing a new reverse-starting point. To address the exposure bias, we introduce a score correction mechanism based on a newly defined score difference. Our approach, which requires no network modifications, is validated across multiple models, datasets, and tasks, achieving state-of-the-art results.Code is at https://github.com/kunzhan/spp