Skip to content
Preprint

Revisiting Diffusion Fine-Tuning for Unsupervised Domain Adaptation

Sep 2026 · 0 citations · 63 references
Computer Science

TL;DR

MUSE (Multi-target UDA-oriented Synthesis with Efficient diffusion fine-tuning), a decoupled adaptation framework that separates source-supervised semantic adaptation from target-specific style adaptation, enabling target-specific generation while avoiding repeated source-guided fine-tuning for each target.

Abstract

Diffusion-based unsupervised domain adaptation (UDA) improves cross-domain transfer by generating target-specific synthetic data for downstream adaptation. Existing methods are largely designed for single-target adaptation: when a model trained on one labeled source domain must be adapted to multiple unlabeled target domains, they typically require separate diffusion fine-tuning for each source--target pair, causing training, storage, and deployment costs to grow with the number of targets. In this paper, we study multi-target data generation for diffusion-based UDA, where a single source-guided diffusion fine-tuning process is reused to generate target-specific synthetic data for multiple target domains. We propose MUSE (Multi-target UDA-oriented Synthesis with Efficient diffusion fine-tuning), a decoupled adaptation framework that separates source-supervised semantic adaptation from target-specific style adaptation. MUSE uses a shared semantic branch updated by labeled source data and target-private style branches specialized to individual target domains, enabling target-specific generation while avoiding repeated source-guided fine-tuning for each target. Experiments on standard UDA benchmarks show that MUSE achieves a stronger accuracy--efficiency trade-off than repeated per-target diffusion adaptation, reducing diffusion fine-tuning cost while improving average target-domain accuracy. The project page is available at https://xuanqi99.github.io/MUSE/.

View source

Similar papers

Open access Aug 2026

Towards source-free domain adaptation: a model-heterogeneous perspective

A Peer-level Heterogeneous Perception Framework is proposed that departs from such paradigms by enabling balanced collaboration between heterogeneous models by introducing an auxiliary domain that is significantly different from the target domain and employ an auxiliary model with the same architecture as the source mo...

Zhi-Ze Wu, Yu-Tao Fu, Huan-Xin Zou et al. · 0 citations
Open access Aug 2026

MDRAN: a multi-source dual-refining adaptation network for source-free domain adaptation guided by vision-language-model

This work proposes a novel criterion, termed Maximum Refinement Decision score (MRD-score), which replaces softmax with median centering to normalize source models’ predictions along both positive and negative axes, thereby harnessing both affirmative and complementary guidance.

Bing-Tao Zhou, Mian Xiang, Qian Ning · 0 citations
Sep 2026

Target-Oriented Multi-Granular Semantic-Aware Network for Source-Free Domain Adaptation.

Source-free Domain Adaptation (SFDA) realizes cross-domain transfer by adapting a pretrained source model to an unlabeled target domain without accessing data in the source domain. Existing SFDA methods mainly develop subdomain-distribution-oriented framework to extract the fine-grained communal semantics from the targ...

Zefeng Zheng, Shaohua Teng, Luyao Teng et al. · 0 citations
Preprint Sep 2026

TC-ADA: One-Shot Active Domain Adaptation for Semantic Segmentation

Target-Calibrated Active Domain Adaptation (TC-ADA) is developed as a joint design of complete-image acquisition and target-calibrated adaptation that selects and densely annotates a fixed target subset in a single round, followed by uninterrupted adaptation.

Wei-Hao Yan, Ye-Qiang Qian, Yue-Yuan Li et al. · 0 citations
Conference Aug 2026

Hierarchically Decoupled Prompt Tuning with Cross-Domain Prediction Alignment for Unsupervised Domain Adaptation

Prompt tuning offers an efficient way to adapt frozen vision-language models to unsupervised domain adaptation (UDA). Existing prompt-based UDA methods usually rely on domain-aware prompts, but they often couple category semantics and domain correction throughout the prompt hierarchy. This coupling can introduce redund...

Kai-Bo Zhou, Xiao-Feng Zhou, Huan-Yang Tong et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.