paper-with-me

Papers

Pluggable Pruning with Contiguous Layer Distillation for Diffusion Transformers

2025-11-20 · Jian Ma, Qirong Peng, Xujie Zhu, Peixing Xie, Chen Chen, Haonan Lu arxiv

Diffusion Transformers (DiTs) have shown exceptional performance in image generation, yet their large parameter counts incur high computational costs, impeding deployment in resource-constrained settings. To address this, we propose Pluggable Pruning with Contiguous Layer Distillation (PPCL), a flexible structured pruning framework specifically designed for DiT architectures. First, we identify redundant layer intervals through a linear probing mechanism combined with the first-order differential trend analysis of similarity metrics. Subsequently, we propose a plug-and-play teacher-student alternating distillation scheme tailored to integrate depth-wise and width-wise pruning within a single training phase. This distillation framework enables flexible knowledge transfer across diverse pruning ratios, eliminating the need for per-configuration retraining. Extensive experiments on multiple Multi-Modal Diffusion Transformer architecture models demonstrate that PPCL achieves a 50\% reduction in parameter count compared to the full model, with less than 3\% degradation in key objective metrics. Notably, our method maintains high-quality image generation capabilities while achieving higher compression ratios, rendering it well-suited for resource-constrained environments. The open-source code, checkpoints for PPCL can be found at the following link: https://github.com/OPPO-Mente-Lab/Qwen-Image-Pruning.

📄 PDF Abstract BibTeX arXiv:2511.16156

Code (0)

등록된 구현이 없습니다.

Tasks

Image Generation

Similar Papers 제목 키워드 기반

LAPTOP-Diff: Layer Pruning and Normalized Distillation for Compressing Diffusion Models

2024-04-17 · Dingkun Zhang, Sijia Li, Chen Chen, Qingsong Xie 외

In the era of AIGC, the demand for low-budget or even on-device applications of diffusion models emerged. In terms of compressing the Stable Diffusion models (SDMs), several approaches have been proposed, and most of the…

Knowledge Distillation

CIVIC: End-to-End Sequence Compactness for Efficient Vision-Language Models

2026-05-27 · Fengze Yang, Bo Yu, Xuewen Luo, Cathy Liu 외 arxiv

Vision-Language Models (VLMs) face severe memory and latency bottlenecks due to high-resolution visual tokens. While current token reduction methods theoretically save FLOPs, post-hoc pruning introduces structural overhe…

Multimodal ReasoningVisual Grounding

Bridging Diffusion Pruning and Step Distillation with Teacher-Aligned Repair

2026-07-07 · Jincheng Ying, Li Wenlin, Minghui Xu, Yinhao Xiao arxiv

Diffusion models generate high-quality images, but their inference cost comes from two sources: large denoising networks and repeated denoising steps. Existing compression pipelines usually attack these costs separately.…

Dense2MoE: Restructuring Diffusion Transformer to MoE for Efficient Text-to-Image Generation

2025-10-10 · Youwei Zheng, Yuxi Ren, Xin Xia, Xuefeng Xiao 외 arxiv

Diffusion Transformer (DiT) has demonstrated remarkable performance in text-to-image generation; however, its large parameter size results in substantial inference overhead. Existing parameter compression methods primari…

Text-to-Image GenerationKnowledge Distillation

Structured Pruning Learns Compact and Accurate Models

2022-04-01 · ACL 2022 5 · Mengzhou Xia, Zexuan Zhong, Danqi Chen

The growing size of neural language models has led to increased attention in model compression. The two predominant approaches are pruning, which gradually removes weights from a pre-trained model, and distillation, whic…

Model Compression