Cover Image

Decouple-Then-Merge: Finetune Diffusion Models as Multi-Task Learning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ma, Qianli, Ning, Xuefei, Liu, Dongrui, Niu, Li, Zhang, Linfeng
Format:	Preprint
Published:	2024
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Accelerating Diffusion Models with One-to-Many Knowledge Distillation
by: Zhang, Linfeng, et al.
Published: (2024)

MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
by: Zhao, Tianchen, et al.
Published: (2024)

Single-Input Multi-Output Model Merging: Leveraging Foundation Models for Dense Multi-Task Learning
by: Giraldo, Juan Garcia, et al.
Published: (2025)

Focused Forcing: Content-Aware Per-Frame KV Selection for Efficient Autoregressive Video Diffusion
by: Cai, Peiliang, et al.
Published: (2026)

Representation Surgery for Multi-Task Model Merging
by: Yang, Enneng, et al.
Published: (2024)

Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
by: Lee, Dong Hoon, et al.
Published: (2024)

Separate-and-Enhance: Compositional Finetuning for Text2Image Diffusion Models
by: Bao, Zhipeng, et al.
Published: (2023)

Towards Unified Modeling in Federated Multi-Task Learning via Subspace Decoupling
by: Wei, Yipan, et al.
Published: (2025)

LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint
by: Ma, Qianli, et al.
Published: (2025)

From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers
by: Liu, Jiacheng, et al.
Published: (2025)

EmotiveTalk: Expressive Talking Head Generation through Audio Information Decoupling and Emotional Video Diffusion
by: Wang, Haotian, et al.
Published: (2024)

Efficient Diffusion as Low Light Enhancer
by: Lan, Guanzhou, et al.
Published: (2024)

SurgeryV2: Bridging the Gap Between Model Merging and Multi-Task Learning with Deep Representation Surgery
by: Yang, Enneng, et al.
Published: (2024)

OrthoDiffusion: A Generalizable Multi-Task Diffusion Foundation Model for Musculoskeletal MRI Interpretation
by: Lan, Tian, et al.
Published: (2026)

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation
by: Ma, Zehong, et al.
Published: (2025)

MaMe & MaRe: Matrix-Based Token Merging and Restoration for Efficient Visual Perception and Synthesis
by: Huo, Simin, et al.
Published: (2026)

Upcycling Text-to-Image Diffusion Models for Multi-Task Capabilities
by: Chavhan, Ruchika, et al.
Published: (2025)

DeepMerge: Deep-Learning-Based Region-Merging for Image Segmentation
by: Lv, Xianwei, et al.
Published: (2023)

DDT: Decoupled Diffusion Transformer
by: Wang, Shuai, et al.
Published: (2025)

Model Collapse in the Self-Consuming Chain of Diffusion Finetuning: A Novel Perspective from Quantitative Trait Modeling
by: Yoon, Youngseok, et al.
Published: (2024)

Task-Driven Exploration: Decoupling and Inter-Task Feedback for Joint Moment Retrieval and Highlight Detection
by: Yang, Jin, et al.
Published: (2024)

ImagebindDC: Compressing Multi-modal Data with Imagebind-based Condensation
by: Min, Yue, et al.
Published: (2025)

Refine-IQA: Multi-Stage Reinforcement Finetuning for Perceptual Image Quality Assessment
by: Jia, Ziheng, et al.
Published: (2025)

Controlling Text-to-Image Diffusion by Orthogonal Finetuning
by: Qiu, Zeju, et al.
Published: (2023)

Finetuning Text-to-Image Diffusion Models for Fairness
by: Shen, Xudong, et al.
Published: (2023)

InpDiffusion: Image Inpainting Localization via Conditional Diffusion Models
by: Wang, Kai, et al.
Published: (2025)

Unified Batch Normalization: Identifying and Alleviating the Feature Condensation in Batch Normalization and a Unified Framework
by: Wang, Shaobo, et al.
Published: (2023)

Layer-wise Model Merging for Unsupervised Domain Adaptation in Segmentation Tasks
by: Alcover-Couso, Roberto, et al.
Published: (2024)

ProMerge: Prompt and Merge for Unsupervised Instance Segmentation
by: Li, Dylan, et al.
Published: (2024)

PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion
by: Lu, Guansong, et al.
Published: (2023)

Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning
by: Ning, Xin, et al.
Published: (2026)

Degrees of Freedom Matter: Inferring Dynamics from Point Trajectories
by: Zhang, Yan, et al.
Published: (2024)

ProReflow: Progressive Reflow with Decomposed Velocity
by: Ke, Lei, et al.
Published: (2025)

PivotMerge: Bridging Heterogeneous Multimodal Pre-training via Post-Alignment Model Merging
by: Shao, Zibo, et al.
Published: (2026)

Denoising Task Routing for Diffusion Models
by: Park, Byeongjun, et al.
Published: (2023)

Identity Decoupling for Multi-Subject Personalization of Text-to-Image Models
by: Jang, Sangwon, et al.
Published: (2024)

MAVIN: Multi-Action Video Generation with Diffusion Models via Transition Video Infilling
by: Zhang, Bowen, et al.
Published: (2024)

Euler-inspired Decoupling Neural Operator for Efficient Pansharpening
by: Zhu, Anqi, et al.
Published: (2026)

Directing the Narrative: A Finetuning Method for Controlling Coherence and Style in Story Generation
by: Zhang, Jianzhang, et al.
Published: (2026)

Merging without Forgetting: Continual Fusion of Task-Specific Models via Optimal Transport
by: Pan, Zecheng, et al.
Published: (2025)