Zobrazeno 1 - 10
of 39
pro vyhledávání: '"Rout, Litu"'
Autor:
Narasimhan, Sai Shankar, Agarwal, Shubhankar, Rout, Litu, Shakkottai, Sanjay, Chinchali, Sandeep P.
Generating realistic time series samples is crucial for stress-testing models and protecting user privacy by using synthetic data. In engineering and safety-critical applications, these samples must meet certain hard constraints that are domain-speci
Externí odkaz:
http://arxiv.org/abs/2410.12652
Autor:
Rout, Litu, Chen, Yujia, Ruiz, Nataniel, Caramanis, Constantine, Shakkottai, Sanjay, Chu, Wen-Sheng
Generative models transform random noise into images; their inversion aims to transform images back to structured noise for recovery and editing. This paper addresses two key tasks: (i) inversion and (ii) editing of a real image using stochastic equi
Externí odkaz:
http://arxiv.org/abs/2410.10792
Autor:
Rout, Litu, Chen, Yujia, Ruiz, Nataniel, Kumar, Abhishek, Caramanis, Constantine, Shakkottai, Sanjay, Chu, Wen-Sheng
We propose Reference-Based Modulation (RB-Modulation), a new plug-and-play solution for training-free personalization of diffusion models. Existing training-free approaches exhibit difficulties in (a) style extraction from reference images in the abs
Externí odkaz:
http://arxiv.org/abs/2405.17401
Autor:
Rout, Litu, Chen, Yujia, Kumar, Abhishek, Caramanis, Constantine, Shakkottai, Sanjay, Chu, Wen-Sheng
Sampling from the posterior distribution poses a major computational challenge in solving inverse problems using latent diffusion models. Common methods rely on Tweedie's first-order moments, which are known to induce a quality-limiting bias. Existin
Externí odkaz:
http://arxiv.org/abs/2312.00852
Autor:
Rout, Litu, Raoof, Negin, Daras, Giannis, Caramanis, Constantine, Dimakis, Alexandros G., Shakkottai, Sanjay
We present the first framework to solve linear inverse problems leveraging pre-trained latent diffusion models. Previously proposed algorithms (such as DPS and DDRM) only apply to pixel-space diffusion models. We theoretically analyze our algorithm s
Externí odkaz:
http://arxiv.org/abs/2307.00619
This work considers the problem of finding a first-order stationary point of a non-convex function with potentially unbounded smoothness constant using a stochastic gradient oracle. We focus on the class of $(L_0,L_1)$-smooth functions proposed by Zh
Externí odkaz:
http://arxiv.org/abs/2302.06570
We provide a theoretical justification for sample recovery using diffusion based image inpainting in a linear model setting. While most inpainting algorithms require retraining with each new mask, we prove that diffusion based inpainting generalizes
Externí odkaz:
http://arxiv.org/abs/2302.01217
Sliced Wasserstein (SW) distance has been widely used in different application scenarios since it can be scaled to a large number of supports without suffering from the curse of dimensionality. The value of sliced Wasserstein distance is the average
Externí odkaz:
http://arxiv.org/abs/2209.13570
Autor:
Gazdieva, Milena, Rout, Litu, Korotin, Alexander, Kravchenko, Andrey, Filippov, Alexander, Burnaev, Evgeny
Real-world image super-resolution (SR) tasks often do not have paired datasets, which limits the application of supervised techniques. As a result, the tasks are usually approached by unpaired techniques based on Generative Adversarial Networks (GANs
Externí odkaz:
http://arxiv.org/abs/2202.01116
With the discovery of Wasserstein GANs, Optimal Transport (OT) has become a powerful tool for large-scale generative modeling tasks. In these tasks, OT cost is typically used as the loss for training GANs. In contrast to this approach, we show that t
Externí odkaz:
http://arxiv.org/abs/2110.02999