KDC-Diff: A Latent-Aware Diffusion Model with Knowledge Retention for Memory-Efficient Image Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Borno, Md. Naimur Asif, Shovon, Md Sakib Hossain, Al-Moisheer, Asmaa Soliman, Moni, Mohammad Ali
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912616417329152
author Borno, Md. Naimur Asif
Shovon, Md Sakib Hossain
Al-Moisheer, Asmaa Soliman
Moni, Mohammad Ali
author_facet Borno, Md. Naimur Asif
Shovon, Md Sakib Hossain
Al-Moisheer, Asmaa Soliman
Moni, Mohammad Ali
contents The growing adoption of generative AI in real-world applications has exposed a critical bottleneck in the computational demands of diffusion-based text-to-image models. In this work, we propose KDC-Diff, a novel and scalable generative framework designed to significantly reduce computational overhead while maintaining high performance. At its core, KDC-Diff designs a structurally streamlined U-Net with a dual-layered knowledge distillation strategy to transfer semantic and structural representations from a larger teacher model. Moreover, a latent-space replay-based continual learning mechanism is incorporated to ensure stable generative performance across sequential tasks. Evaluated on benchmark datasets, our model demonstrates strong performance across FID, CLIP, KID, and LPIPS metrics while achieving substantial reductions in parameter count, inference time, and FLOPs. KDC-Diff offers a practical, lightweight, and generalizable solution for deploying diffusion models in low-resource environments, making it well-suited for the next generation of intelligent and resource-aware computing systems.
format Preprint
id arxiv_https___arxiv_org_abs_2505_06995
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle KDC-Diff: A Latent-Aware Diffusion Model with Knowledge Retention for Memory-Efficient Image Generation
Borno, Md. Naimur Asif
Shovon, Md Sakib Hossain
Al-Moisheer, Asmaa Soliman
Moni, Mohammad Ali
Computer Vision and Pattern Recognition
The growing adoption of generative AI in real-world applications has exposed a critical bottleneck in the computational demands of diffusion-based text-to-image models. In this work, we propose KDC-Diff, a novel and scalable generative framework designed to significantly reduce computational overhead while maintaining high performance. At its core, KDC-Diff designs a structurally streamlined U-Net with a dual-layered knowledge distillation strategy to transfer semantic and structural representations from a larger teacher model. Moreover, a latent-space replay-based continual learning mechanism is incorporated to ensure stable generative performance across sequential tasks. Evaluated on benchmark datasets, our model demonstrates strong performance across FID, CLIP, KID, and LPIPS metrics while achieving substantial reductions in parameter count, inference time, and FLOPs. KDC-Diff offers a practical, lightweight, and generalizable solution for deploying diffusion models in low-resource environments, making it well-suited for the next generation of intelligent and resource-aware computing systems.
title KDC-Diff: A Latent-Aware Diffusion Model with Knowledge Retention for Memory-Efficient Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.06995