mlsanityData to 7 Oct 2026

Source Separation with Deep Generative Priors

arXiv 2002.07942 ↗First posted 19 Feb 2020cs.LG stat.MLcited by 35

Not on the map. Why?

Abstract (arXiv)

Despite substantial progress in signal source separation, results for richly structured data continue to contain perceptible artifacts. In contrast, recent deep generative models can produce authentic samples in a variety of domains that are indistinguishable from samples of the data distribution. This paper introduces a Bayesian approach to source separation that uses generative models as priors over the components of a mixture of sources, and noise-annealed Langevin dynamics to sample from the posterior distribution of sources given a mixture. This decouples the source separation problem from generative modeling, enabling us to directly use cutting-edge generative models as priors. The method achieves state-of-the-art performance for MNIST digit separation. We introduce new methodology for evaluating separation quality on richer datasets, providing quantitative evaluation of separation results on CIFAR-10. We also provide qualitative results on LSUN.

Key facts. Cited by 35 papers in the map (data through 7 Oct 2026). First posted on arXiv 19 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 6; 2023: 8; 2024: 6; 2025: 6; 2026: 3.

35 papers.

Sort:

By area:

  1. 2021Gotta Go Fast When Generating Data with Score-Based Modelscited by 218
  2. 2023Multi-Source Diffusion Models for Simultaneous Music Generation and Separationcited by 63
  3. 2022Denoising Likelihood Score Matching for Conditional Score-based Data Generationcited by 45
  4. 2022Diffusion-based Generative Speech Source Separationcited by 41
  5. 2021Parallel and Flexible Sampling from Autoregressive Models via Langevin Dynamicscited by 19
  6. 2021Blind Image Decompositioncited by 18
  7. 2022Astronomia ex machina: a history, primer, and outlook on neural networks in astronomycited by 17
  8. 2021Detector-Free Weakly Supervised Grounding by Separationcited by 16
  9. 2023Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separationcited by 14
  10. 2024COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representationscited by 14
  11. 2024Generalized Multi-Source Inference for Text Conditioned Music Diffusion Modelscited by 13
  12. 2024Simultaneous Music Separation and Generation Using Multi-Track Latent Diffusion Modelscited by 13
23 more
  1. 2023Latent Autoregressive Source Separationcited by 12
  2. 2022One-shot Generative Prior in Hankel-k-space for Parallel Imaging Reconstructioncited by 11
  3. 2021Wavelet Transform-assisted Adaptive Generative Modeling for Colorizationcited by 8
  4. 2025Source Separation by Flow Matchingcited by 8
  5. 2023Score-based Source Separation with Applications to Digital Communication Signalscited by 7
  6. 2025ArrayDPS: Unsupervised Blind Speech Separation with a Diffusion Priorcited by 7
  7. 2025MGE-LDM: Joint Latent Diffusion for Simultaneous Music Generation and Source Extractioncited by 4
  8. 2025ZeroSep: Separate Anything in Audio with Zero Trainingcited by 4
  9. 2023Diffusion-based Signal Refiner for Speech Enhancement and Separationcited by 3
  10. 2025Unsupervised Single-Channel Audio Separation with Diffusion Source Priorscited by 3
  11. 2021Unsupervised Source Separation via Bayesian Inference in the Latent Domaincited by 2
  12. 2022Learning and controlling the source-filter representation of speech with a variational autoencodercited by 2
  13. 2026The Radio-Frequency Transformer for Signal Separationcited by 2
  14. 2026StrADiff: A Structured Source-Wise Adaptive Diffusion Framework for Linear and Nonlinear Blind Source Separationcited by 2
  15. 2023Blind Source Separation of Single-Channel Mixtures via Multi-Encoder Autoencoderscited by 1
  16. 2022Semi-Blind Source Separation with Learned Constraintscited by 0
  17. 2023Adversarial Generative NMF for Single Channel Source Separationcited by 0
  18. 2023Mixup-Privacy: A simple yet effective approach for privacy-preserving segmentationcited by 0
  19. 2024Weighted-Sum of Gaussian Process Latent Variable Modelscited by 0
  20. 2024Maximum Discrepancy Generative Regularization and Non-Negative Matrix Factorization for Single Channel Source Separationcited by 0
  21. 2024Source Separation of Multi-source Raw Music using a Residual Quantized Variational Autoencodercited by 0
  22. 2025Disentanglement of Sources in a Multi-Stream Variational Autoencodercited by 0
  23. 2026SURF: Separation via Unsupervised Remixing Flowcited by 0