Picture for Piyush Rai

Piyush Rai

University of Utah

RISSOLE: Parameter-efficient Diffusion Models via Block-wise Generation and Retrieval-Guidance

Add code
Sep 02, 2024
Viaarxiv icon

Robust Black-box Testing of Deep Neural Networks using Co-Domain Coverage

Add code
Aug 13, 2024
Viaarxiv icon

VERSE: Virtual-Gradient Aware Streaming Lifelong Learning with Anytime Inference

Add code
Sep 15, 2023
Viaarxiv icon

A Probabilistic Framework for Lifelong Test-Time Adaptation

Add code
Dec 19, 2022
Viaarxiv icon

Novel Class Discovery without Forgetting

Add code
Jul 21, 2022
Figure 1 for Novel Class Discovery without Forgetting
Figure 2 for Novel Class Discovery without Forgetting
Figure 3 for Novel Class Discovery without Forgetting
Figure 4 for Novel Class Discovery without Forgetting
Viaarxiv icon

Bayesian Federated Learning via Predictive Distribution Distillation

Add code
Jun 15, 2022
Figure 1 for Bayesian Federated Learning via Predictive Distribution Distillation
Figure 2 for Bayesian Federated Learning via Predictive Distribution Distillation
Figure 3 for Bayesian Federated Learning via Predictive Distribution Distillation
Figure 4 for Bayesian Federated Learning via Predictive Distribution Distillation
Viaarxiv icon

Spacing Loss for Discovering Novel Categories

Add code
Apr 22, 2022
Figure 1 for Spacing Loss for Discovering Novel Categories
Figure 2 for Spacing Loss for Discovering Novel Categories
Figure 3 for Spacing Loss for Discovering Novel Categories
Viaarxiv icon

Semi-Supervised Super-Resolution

Add code
Apr 19, 2022
Figure 1 for Semi-Supervised Super-Resolution
Figure 2 for Semi-Supervised Super-Resolution
Figure 3 for Semi-Supervised Super-Resolution
Figure 4 for Semi-Supervised Super-Resolution
Viaarxiv icon

DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents

Add code
Jan 02, 2022
Figure 1 for DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents
Figure 2 for DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents
Figure 3 for DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents
Figure 4 for DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents
Viaarxiv icon

SITA: Single Image Test-time Adaptation

Add code
Dec 08, 2021
Figure 1 for SITA: Single Image Test-time Adaptation
Figure 2 for SITA: Single Image Test-time Adaptation
Figure 3 for SITA: Single Image Test-time Adaptation
Figure 4 for SITA: Single Image Test-time Adaptation
Viaarxiv icon