Noise-Robust Conditional Flow Matching: Generating Clean Samples from Noisy Datasets

• URL: http://arxiv.org/abs/2608.00064v2
• Date: Tue, 04 Aug 2026 14:16:56 GMT
• Title: Noise-Robust Conditional Flow Matching: Generating Clean Samples from Noisy Datasets
• Abstract summary: Conditional flow matching (CFM) combines a simple regression objective with stable training, efficient sampling, and strong image-generation performance.<n>We introduce Noise-Robust Conditional Flow Matching (NR-CFM), an unconditional generator that learns from one corrupted observation per image.
• Score: 3.998139009044269
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generative models learn the statistical properties of their training data, so high-quality generation depends on clean and representative datasets. In scientific imaging, acquisition often yields noisy measurements, while collecting clean references can be costly, impractical or even unattainable. Training directly on these measurements results in a model that reproduces the corrupted data. This can be circumvented by learning the clean population distribution directly from the noisy data. Conditional flow matching (CFM) combines a simple regression objective with stable training, efficient sampling, and strong image-generation performance, making it a natural framework for this setting. We introduce Noise-Robust Conditional Flow Matching (NR-CFM), an unconditional generator that learns from one corrupted observation per image. NR-CFM provides a closed-form clean endpoint correction for additive white Gaussian noise and learns a data-driven correction for general Gaussian corruptions with more complex covariance structure. Across the evaluated corruption settings, NR-CFM outperforms NR-GAN in most cases and remains competitive with Ambient Diffusion in the high-noise regime. We further evaluate NR-CFM on scientific data at signal-to-noise ratios as low as $0.001$, where it generates plausible particle images from severely corrupted measurements.

Related papers

• Interleaved Noise Injection Improves Clean, Corrupted, and OOD Performance [1.6567880228735357]
  We present a theoretical analysis of noise injection, which confirms that corruption by impulse noise approximates a Jacobian regularization.<n>We show that the effect of interleaved noise injection stems from penalizing the failure modes encouraged by the inductive bias of the models.<n>Interleaved noise injection is therefore an effective tool to improve the test performance on clean, noisy, and out-of-distribution data at essentially zero computational cost.
  arXiv  Detail & Related papers  (2026-07-16T01:25:44Z)
• Realistic noise synthesis reduces bias and improves tissue microstructure estimation with supervised machine learning [0.2155083987188202]
  We investigate the impact of discrepancies between noise characteristics of simulated and acquired signals on microstructure parameter estimation.<n>We propose a realistic noise synthesis (RNS) framework to mitigate it.<n>RNS incorporates both the Rician expectation and the effective post-processing noise variance into simulated training signals.
  arXiv  Detail & Related papers  (2026-06-01T10:35:33Z)
• Memory-Distilled Selection for Noise-Robust Anomaly Detection [17.179366730533438]
  Anomaly detection (AD) under data contamination is critical for deploying unsupervised defect detection in industrial environments.<n>We propose Memory-Distilled Selection (MeDS), a training algorithm based on data selection.<n>MeDS constructs an ensemble of partial memories via random subsampling, where the resulting sparsity acts as a low-pass filter.
  arXiv  Detail & Related papers  (2026-05-26T08:13:41Z)
• Mitigating the Noise Shift for Denoising Generative Models via Noise Awareness Guidance [54.88271057438763]
  Noise Awareness Guidance (NAG) is a correction method that explicitly steers sampling trajectories to remain consistent with the pre-defined noise schedule.<n>NAG consistently mitigates noise shift and substantially improves the generation quality of mainstream diffusion models.
  arXiv  Detail & Related papers  (2025-10-14T13:31:34Z)
• Low-Field Magnetic Resonance Image Quality Enhancement using a Conditional Flow Matching Model [0.3683202928838613]
  conditional flow matching (CFM) learns a continuous flow between a noise distribution and target data distributions.<n>Our framework is designed to reconstruct high-field-like MR images from their corresponding low-field inputs.<n>Experiments demonstrate that CFM achieves state-of-the-art performance, but also generalizes robustly to both in-distribution and out-of-distribution data.
  arXiv  Detail & Related papers  (2025-10-14T11:41:27Z)
• On the Shape of Latent Variables in a Denoising VAE-MoG: A Posterior Sampling-Based Study [51.56484100374058]
  We explore the latent space of a denoising variational autoencoder with a mixture-of-Gaussians prior (VAE-MoG)<n>To evaluate how well the model captures the underlying structure, we use Hamiltonian Monte Carlo (HMC) to draw posterior samples conditioned on clean inputs, and compare them to the encoder's outputs from noisy data.<n>Although the model reconstructs signals accurately, statistical comparisons reveal a clear mismatch in the latent space.
  arXiv  Detail & Related papers  (2025-09-29T18:33:09Z)
• Denoising Score Distillation: From Noisy Diffusion Pretraining to One-Step High-Quality Generation [82.39763984380625]
  We introduce denoising score distillation (DSD), a surprisingly effective and novel approach for training high-quality generative models from low-quality data.<n>DSD pretrains a diffusion model exclusively on noisy, corrupted samples and then distills it into a one-step generator capable of producing refined, clean outputs.
  arXiv  Detail & Related papers  (2025-03-10T17:44:46Z)
• GSURE-Based Diffusion Model Training with Corrupted Data [35.56267114494076]
  We propose a novel training technique for generative diffusion models based only on corrupted data. We demonstrate our technique on face images as well as Magnetic Resonance Imaging (MRI)
  arXiv  Detail & Related papers  (2023-05-22T15:27:20Z)
• Adaptive Multi-View ICA: Estimation of noise levels for optimal inference [65.94843987207445]
  Adaptive multiView ICA (AVICA) is a noisy ICA model where each view is a linear mixture of shared independent sources with additive noise on the sources. On synthetic data, AVICA yields better sources estimates than other group ICA methods thanks to its explicit MMSE estimator. On real magnetoencephalograpy (MEG) data, we provide evidence that the decomposition is less sensitive to sampling noise and that the noise variance estimates are biologically plausible.
  arXiv  Detail & Related papers  (2021-02-22T13:10:12Z)
• Learning Energy-Based Models by Diffusion Recovery Likelihood [61.069760183331745]
  We present a diffusion recovery likelihood method to tractably learn and sample from a sequence of energy-based models. After training, synthesized images can be generated by the sampling process that initializes from Gaussian white noise distribution. On unconditional CIFAR-10 our method achieves FID 9.58 and inception score 8.30, superior to the majority of GANs.
  arXiv  Detail & Related papers  (2020-12-15T07:09:02Z)
• Residual-driven Fuzzy C-Means Clustering for Image Segmentation [152.609322951917]
  We elaborate on residual-driven Fuzzy C-Means (FCM) for image segmentation. Built on this framework, we present a weighted $ell_2$-norm fidelity term by weighting mixed noise distribution. The results demonstrate the superior effectiveness and efficiency of the proposed algorithm over existing FCM-related algorithms.
  arXiv  Detail & Related papers  (2020-04-15T15:46:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.