Reinforcement learning based post training of diffusion models, such as Denoising Diffusion Policy Optimization (DDPO), optimizes a reverse diffusion process under a reward function. However, current approaches to reward optimizations do so at the cost of diversity and quality. In this paper, we pro
iADD: Improving Alignment and Diversity in Diffusion Policy Optimization
Researchers propose iADD, a method for reinforcement-learning post-training of diffusion models that aims to improve alignment with reward functions without sacrificing output diversity, addressing a limitation of Denoising Diffusion Policy Optimization (DDPO). DDPO optimizes a reverse diffusion process under a reward function, but the paper states current reward-optimization approaches achieve this at the cost of diversity and quality. The work targets the trade-off between reward alignment and sample diversity in diffusion policy optimization.
iADD is presented as a diffusion policy-optimization method that improves reward alignment while preserving diversity and quality, which DDPO-style reward optimization sacrifices.
Run your AI side-project on zahid.host
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.