activity
20242026
collaborators
Showing cs.CVShow all

9 papers · 1 filter

cs.CV2026

Score-Control for Hallucination Reduction in Diffusion Models

Mahesh Bhosale, Naresh Kumar Devulapally, Abdul Wasi +3

Diffusion models have emerged as the backbone of modern generative AI, powering advances in vision, language, audio and other modalities. Despite their success, they suffer from ha…

cs.CV2026

ManifoldGD: Training-Free Hierarchical Manifold Guidance for Diffusion-Based Dataset Distillation

Ayush Roy, Wei-Yang Alex Lee, Rudrasis Chakraborty +1

In recent times, large datasets hinder efficient model training while also containing redundant concepts. Dataset distillation aims to synthesize compact datasets that preserve the…

cs.CV2026

Is Exchangeability better than I.I.D to handle Data Distribution Shifts while Pooling Data for Data-scarce Medical image segmentation?

Ayush Roy, Samin Enam, Jun Xia +2

Data scarcity is a major challenge in medical imaging, particularly for deep learning models. While data pooling (combining datasets from multiple sources) and data addition (addin…

cs.CV2026

Forget Less by Learning Together through Concept Consolidation

Arjun Ramesh Kaushik, Naresh Kumar Devulapally, Vishnu Suresh Lokhande +2

Custom Diffusion Models (CDMs) have gained significant attention due to their remarkable ability to personalize generative processes. However, existing CDMs suffer from catastrophi…

cs.CV2026

Forget Less by Learning from Parents Through Hierarchical Relationships

Arjun Ramesh Kaushik, Naresh Kumar Devulapally, Vishnu Suresh Lokhande +2

Custom Diffusion Models (CDMs) offer impressive capabilities for personalization in generative modeling, yet they remain vulnerable to catastrophic forgetting when learning new con…

cs.CV2025

Latent Diffusion Unlearning: Protecting Against Unauthorized Personalization Through Trajectory Shifted Perturbations

Naresh Kumar Devulapally, Shruti Agarwal, Tejas Gokhale +1

Text-to-image diffusion models have demonstrated remarkable effectiveness in rapid and high-fidelity personalization, even when provided with only a few user images. However, the e…