The paper introduces a novel framework called Plug-in Diffusion Model for Recommendation (PDRec), which leverages diffusion models to enhance sequential recommendation (SR) systems. PDRec addresses the data sparsity issue by fully utilizing the diffusion-generated user preferences on all items, not just the highest-scored items. The framework includes three main components: Historical Behavior Reweighting (HBR), Diffusion-based Positive Augmentation (DPA), and Noise-Free Negative Sampling (NNS). HBR reweights historical behaviors to identify high-quality interactions, DPA uses unobserved items with high diffusion-based preferences as potential positive samples, and NNS selects stable negative samples to mitigate false negatives. Extensive experiments on four datasets and three base SR models demonstrate that PDRec significantly outperforms state-of-the-art baselines, showing its effectiveness and universality across different recommendation scenarios. The code for PDRec is available at https://github.com/hulkima/PDRec.The paper introduces a novel framework called Plug-in Diffusion Model for Recommendation (PDRec), which leverages diffusion models to enhance sequential recommendation (SR) systems. PDRec addresses the data sparsity issue by fully utilizing the diffusion-generated user preferences on all items, not just the highest-scored items. The framework includes three main components: Historical Behavior Reweighting (HBR), Diffusion-based Positive Augmentation (DPA), and Noise-Free Negative Sampling (NNS). HBR reweights historical behaviors to identify high-quality interactions, DPA uses unobserved items with high diffusion-based preferences as potential positive samples, and NNS selects stable negative samples to mitigate false negatives. Extensive experiments on four datasets and three base SR models demonstrate that PDRec significantly outperforms state-of-the-art baselines, showing its effectiveness and universality across different recommendation scenarios. The code for PDRec is available at https://github.com/hulkima/PDRec.