{"ID":3085596,"CreatedAt":"2026-06-05T06:46:15.197025399Z","UpdatedAt":"2026-06-06T05:44:34.749899951Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.05239","arxiv_id":"2606.05239","title":"HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation","abstract":"Diffusion models have demonstrated strong performance in time series modeling due to their ability to progressively capture complex data distributions through iterative denoising. However, existing approaches struggle with frequency-sensitive denoising, high-frequency reconstruction and balancing global trends with local dynamics. To address these limitations, we propose \\textbf{HyFAD}, a \\textbf{Hy}brid time-frequency \\textbf{D}iffusion model with \\textbf{F}requency-\\textbf{A}ware embedding for time series imputation. Built upon the DDPM paradigm, HyFAD adopts a coupled time-frequency diffusion framework, in which the reverse denoising proceeds sequentially from the time domain to the frequency domain, enabling coarse-to-fine generation. Specifically, the time-domain diffusion process captures low-frequency global trends, while the frequency-domain diffusion process refines high-frequency spectral components. We further introduce a frequency-aware step embedding that exploits the relationship between diffusion steps and spectral components, providing step-dependent spectral guidance and facilitates more accurate band-wise reconstruction. Extensive experiments on multiple benchmark datasets demonstrate that HyFAD achieves state-of-the-art performance. Our source code is available at https://github.com/hongfangao/HyFAD.","short_abstract":"Diffusion models have demonstrated strong performance in time series modeling due to their ability to progressively capture complex data distributions through iterative denoising. However, existing approaches struggle with frequency-sensitive denoising, high-frequency reconstruction and balancing global trends with loc...","url_abs":"https://arxiv.org/abs/2606.05239","url_pdf":"https://arxiv.org/pdf/2606.05239v1","authors":"[\"Hongfan Gao\",\"Wangmeng Shen\",\"Bin Yang\",\"Jilin Hu\"]","published":"2026-06-03T06:03:37Z","proceeding":"stat.ML","tasks":"[\"stat.ML\",\"cs.LG\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":612869,"CreatedAt":"2026-06-05T06:46:15.197025399Z","UpdatedAt":"2026-06-05T06:46:15.197025399Z","DeletedAt":null,"paper_id":3085596,"paper_url":"https://arxiv.org/abs/2606.05239","paper_title":"HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation","repo_url":"https://github.com/hongfangao/HyFAD","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}