arXiv subjects

Darren Lo

Publications and source records attributed to Darren Lo.

4 recordsLinked to original sources

arXiv2025

SFBD-OMNI: Bridge models for lossy measurement restoration with limited clean samples

In many real-world scenarios, obtaining fully observed samples is prohibitively expensive or even infeasible, while partial and noisy observations are comparatively easy to collect. In this work, we study distribution restoration with abundant noisy samples, assuming the corruption process is available as a black-box generator. We show that this task can be framed as a one-sided entropic optimal transport problem and solved via an EM-like algorithm. We further provide a test criterion to determine whether the true underlying distribution is recoverable under per-sample information loss, and show that in otherwise unrecoverable cases, a small number of clean samples can render the distribution largely recoverable. Building on these insights, we introduce SFBD-OMNI, a bridge model-based framework that maps corrupted sample distributions to the ground-truth distribution. Our method generalizes Stochastic Forward-Backward Deconvolution (SFBD; Lu et al., 2025) to handle arbitrary measurement models beyond Gaussian corruption. Experiments across benchmark datasets and diverse measurement settings demonstrate significant improvements in both qualitative and quantitative performance.

cs.LG↗

arXiv2025

Composite Classifier-Free Guidance for Multi-Modal Conditioning in Wind Dynamics Super-Resolution

Various weather modelling problems (e.g., weather forecasting, optimizing turbine placements, etc.) require ample access to high-resolution, highly accurate wind data. Acquiring such high-resolution wind data, however, remains a challenging and expensive endeavour. Traditional reconstruction approaches are typically either cost-effective or accurate, but not both. Deep learning methods, including diffusion models, have been proposed to resolve this trade-off by leveraging advances in natural image super-resolution. Wind data, however, is distinct from natural images, and wind super-resolvers often use upwards of 10 input channels, significantly more than the usual 3-channel RGB inputs in natural images. To better leverage a large number of conditioning variables in diffusion models, we present a generalization of classifier-free guidance (CFG) to multiple conditioning inputs. Our novel composite classifier-free guidance (CCFG) can be dropped into any pre-trained diffusion model trained with standard CFG dropout. We demonstrate that CCFG outputs are higher-fidelity than those from CFG on wind super-resolution tasks. We present WindDM, a diffusion model trained for industrial-scale wind dynamics reconstruction and leveraging CCFG. WindDM achieves state-of-the-art reconstruction quality among deep learning models and costs up to $1000\times$ less than classical methods.

cs.LG↗

arXiv2025

SFBD Flow: A Continuous-Optimization Framework for Training Diffusion Models with Noisy Samples

Diffusion models achieve strong generative performance but often rely on large datasets that may include sensitive content. This challenge is compounded by the models' tendency to memorize training data, raising privacy concerns. SFBD (Lu et al., 2025) addresses this by training on corrupted data and using limited clean samples to capture local structure and improve convergence. However, its iterative denoising and fine-tuning loop requires manual coordination, making it burdensome to implement. We reinterpret SFBD as an alternating projection algorithm and introduce a continuous variant, SFBD flow, that removes the need for alternating steps. We further show its connection to consistency constraint-based methods, and demonstrate that its practical instantiation, Online SFBD, consistently outperforms strong baselines across benchmarks.

cs.LG↗

arXiv2021

The Evolution of Flow and Mass Transport in 3D Confined Cavities

Flow in channels and ducts with adjoining cavities are common in natural and engineered systems. Here we report numerical and experimental results of 3D confined cavity flow, identifying critical conditions in the recirculating flow formation and mass transport over a range of channel flow properties ($0.1\leq Re \leq 300$) and cavity aspect ratio ($0.1 \leq H/X_{s} \leq1$). In contrast to 2D systems, a mass flux boundary is not formed in 3D confined cavity-channel flow. Streamlines directly enter a recirculating vortex in the cavity and exit to the main channel leading to an exponential increase in the cavity mass flux when the recirculating vortex fills the cavity volume. These findings extend our understanding of flow entry and exit in cavities and suggest conditions where convective mass transport into and out of cavities would be amplified and vortex particle capture reduced.

physics.flu-dyn↗