9 papers
Next-Dense-Stride Prediction for Multimodal Autoregressive Visual Modeling
Chicago Y. Park, Jialin Mao, Xiaojian Xu +3
We introduce DenseAR, a new generative paradigm that reformulates autoregressive image generation as coarse-to-fine next-dense-stride prediction using a compact single-scale tokeni…
NullFlow: One-Step Generative Reconstruction
Xiao Shi, Edward P. Chandler, Chicago Y. Park +2
We propose NullFlow, a principled framework for one-step generative image reconstruction. Our key idea is to confine the generative flow to a measurement-consistent subspace. Becau…
Stochastic Generative Plug-and-Play Priors
Chicago Y. Park, Edward P. Chandler, Yuyang Hu +4
Plug-and-play (PnP) methods are widely used for solving imaging inverse problems by incorporating a denoiser into optimization algorithms. Score-based diffusion models (SBDMs) have…
Deep Parameter Interpolation for Scalar Conditioning
Chicago Y. Park, Michael T. McCann, Cristina Garcia-Cardona +2
We propose deep parameter interpolation (DPI), a general-purpose method for transforming an existing deep neural network architecture into one that accepts an additional scalar inp…
Measurement Score-Based MRI Reconstruction with Automatic Coil Sensitivity Estimation
Tingjun Liu, Chicago Y. Park, Yuyang Hu +2
Diffusion-based inverse problem solvers (DIS) have recently shown outstanding performance in compressed-sensing parallel MRI reconstruction by combining diffusion priors with physi…
Random Walks with Tweedie: A Unified View of Score-Based Diffusion Models
Chicago Y. Park, Michael T. McCann, Cristina Garcia-Cardona +2
We present a concise derivation for several influential score-based diffusion models that relies on only a few textbook results. Diffusion models have recently emerged as powerful…