MAST: Mask-Guided Attention Mass Allocation for Training-Free Multi-Style Transfer

• URL: http://arxiv.org/abs/2604.12281v1
• Date: Tue, 14 Apr 2026 04:47:09 GMT
• Title: MAST: Mask-Guided Attention Mass Allocation for Training-Free Multi-Style Transfer
• Abstract summary: We propose MAST (Mask-Guided Attention Mass Allocation for Training-Free Multi-Style Transfer), a training-free framework for style transfer.<n>To achieve artifact-free and structure-preserving stylization, MAST integrates four connected modules.
• Score: 6.817047561934744
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Style transfer aims to render a content image with the visual characteristics of a reference style while preserving its underlying semantic layout and structural geometry. While recent diffusion-based models demonstrate strong stylization capabilities by leveraging powerful generative priors and controllable internal representations, they typically assume a single global style. Extending them to multi-style scenarios often leads to boundary artifacts, unstable stylization, and structural inconsistency due to interference between multiple style representations. To overcome these limitations, we propose MAST (Mask-Guided Attention Mass Allocation for Training-Free Multi-Style Transfer), a novel training-free framework that explicitly controls content-style interactions within the diffusion attention mechanism. To achieve artifact-free and structure-preserving stylization, MAST integrates four connected modules. First, Layout-preserving Query Anchoring prevents global layout collapse by firmly anchoring the semantic structure using content queries. Second, Logit-level Attention Mass Allocation deterministically distributes attention probability mass across spatial regions, seamlessly fusing multiple styles without boundary artifacts. Third, Sharpness-aware Temperature Scaling restores the attention sharpness degraded by multi-style expansion. Finally, Discrepancy-aware Detail Injection adaptively compensates for localized high-frequency detail losses by measuring structural discrepancies. Extensive experiments demonstrate that MAST effectively mitigates boundary artifacts and maintains structural consistency, preserving texture fidelity and spatial coherence even as the number of applied styles increases.

Related papers

• Isotropic Embedding Perturbations for Robust Vision Language Encoders [73.92626712781859]
  Aether is a simple plug-in method that applies diffusion-style random perturbations in the embedding space via controlled alpha-mixing.<n>Inspired by feature-space perturbations in language models and image degradation in generative pretraining, Aether induces mild yet effective perturbations.<n>Aether delivers consistent gains over the advanced recipe combining CutMix, Mixup, DropPath, and RandAug.
  arXiv  Detail & Related papers  (2026-09-09T15:08:05Z)
• UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation [65.53694602893042]
  We propose a unified visual conditioning framework that fuses ViT and VAE features before VLM encoding.<n>Experiments on two multi-reference generation benchmarks demonstrate that UniCustom consistently improves subject consistency, instruction following, and compositional fidelity.
  arXiv  Detail & Related papers  (2026-05-12T13:10:05Z)
• UniCSG: Unified High-Fidelity Content-Constrained Style-Driven Generation via Staged Semantic and Frequency Disentanglement [13.265643008768825]
  DiT-based diffusion models often suffer from content-style entanglement, leading to reference-content leakage and unstable generation.<n>We present UniCSG, a unified framework for content-constrained, style-driven generation in both text-guided and reference-guided settings.
  arXiv  Detail & Related papers  (2026-04-20T05:59:20Z)
• EReCu: Pseudo-label Evolution Fusion and Refinement with Multi-Cue Learning for Unsupervised Camouflage Detection [20.892452296357813]
  Unsupervised Camouflaged Object Detection (UCOD) remains a challenging task due to the high intrinsic similarity between target objects and their surroundings.<n>We propose a unified UCOD framework that enhances both the reliability of pseudo-labels and the fidelity of features.<n>Our method achieves state-of-the-art performance, characterized by superior detail perception, robust boundary alignment, and strong generalization.
  arXiv  Detail & Related papers  (2026-03-12T04:10:32Z)
• Deepfake Detection with Multi-Artifact Subspace Fine-Tuning and Selective Layer Masking [11.158258169109907]
  Deepfake detection still faces significant challenges in cross-dataset and real-world complex scenarios.<n>This paper proposes a deepfake detection method based on Multi-Artifact Subspaces and selective layer masks (MASM)<n>MASM explicitly decouples semantic representations from artifact representations and constrains the fitting strength of artifact subspaces.
  arXiv  Detail & Related papers  (2026-01-03T02:33:18Z)
• AnyMS: Bottom-up Attention Decoupling for Layout-guided and Training-free Multi-subject Customization [55.06425570300248]
  We present AnyMS, a training-free framework for layout-guided multi-subject customization.<n>AnyMS leverages three input conditions: text prompt, subject images, and layout constraints.<n>AnyMS achieves state-of-the-art performance, supporting complex compositions and scaling to a larger number of subjects.
  arXiv  Detail & Related papers  (2025-12-29T15:26:25Z)
• MeshMosaic: Scaling Artist Mesh Generation via Local-to-Global Assembly [62.48017648785026]
  We introduce MeshMosaic, a novel local-to-global framework for artist mesh generation that scales to over 100K triangles.<n>We show that MeshMosaic significantly outperforms state-of-the-art methods in both geometric fidelity and user preference.
  arXiv  Detail & Related papers  (2025-09-24T11:02:03Z)
• ICAS: IP Adapter and ControlNet-based Attention Structure for Multi-Subject Style Transfer Optimization [0.0]
  ICAS is a novel framework for efficient and controllable multi-subject style transfer.<n>Our framework ensures faithful global layout preservation alongside accurate local style synthesis.<n>ICAS achieves superior performance in structure preservation, style consistency, and inference efficiency.
  arXiv  Detail & Related papers  (2025-04-17T10:48:11Z)
• FreSca: Scaling in Frequency Space Enhances Diffusion Models [55.75504192166779]
  This paper explores frequency-based control within latent diffusion models.<n>We introduce FreSca, a novel framework that decomposes noise difference into low- and high-frequency components.<n>FreSca operates without any model retraining or architectural change, offering model- and task-agnostic control.
  arXiv  Detail & Related papers  (2025-04-02T22:03:11Z)
• UniVST: A Unified Framework for Training-free Localized Video Style Transfer [102.52552893495475]
  This paper presents UniVST, a unified framework for localized video style transfer based on diffusion models.<n>It operates without the need for training, offering a distinct advantage over existing diffusion methods that transfer style across entire videos.
  arXiv  Detail & Related papers  (2024-10-26T05:28:02Z)
• SSMG: Spatial-Semantic Map Guided Diffusion Model for Free-form Layout-to-Image Generation [68.42476385214785]
  We propose a novel Spatial-Semantic Map Guided (SSMG) diffusion model that adopts the feature map, derived from the layout, as guidance. SSMG achieves superior generation quality with sufficient spatial and semantic controllability compared to previous works. We also propose the Relation-Sensitive Attention (RSA) and Location-Sensitive Attention (LSA) mechanisms.
  arXiv  Detail & Related papers  (2023-08-20T04:09:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.