SHED: Style-Homogenized Embedding Alignment for Domain Generalization

• URL: http://arxiv.org/abs/2605.16973v1
• Date: Sat, 16 May 2026 12:51:38 GMT
• Title: SHED: Style-Homogenized Embedding Alignment for Domain Generalization
• Abstract summary: We propose Style-Homogenized Embedding alignment for Domain-generalization (SHED)<n>SHED aligns style-homogenized embeddings instead of raw representations from encoders in CLIP.<n>Experiments on five benchmarks show SHED achieves state-of-the-art performance.
• Score: 13.548252728962966
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Domain generalization aims to enhance model robustness against unseen domains with embedding distribution shifts. While large-scale vision-language models like CLIP exhibit strong generalization, their direct image-text embedding alignment suffers from inherent information asymmetry: images encode both class semantics and domain-specific styles, whereas text prompts primarily convey basic class cues. This asymmetry hinders generalization to novel domains in realistic scenarios. To address this, we propose Style-Homogenized Embedding alignment for Domain-generalization (SHED), a novel CLIP-based method that aligns style-homogenized embeddings instead of raw representations from encoders in CLIP. During training, SHED removes domain-specific style centroids from both image embeddings computed per source domains and text embeddings which are averaged across diverse prompt templates and stripped of a global centroid. For inference, considering the lack of target domain information, SHED projects diverse textual domain centroids into the visual space and aggregates predictions via membership weighting. Extensive experiments on five benchmarks show SHED achieves state-of-the-art performance, outperforming prior methods significantly (e.g., +4.0\% on DomainNet vs. standard fine-tuning).

Related papers

• LASA: Language-and-Source-Anchored Alignment for Domain Generalized Semantic Segmentation [43.56251979741165]
  Domain Generalization Semantic (DGSS) focuses on generalizing knowledge from labeled source domains to unseen target domains.<n>Style randomization distorts the underlying feature manifold due to its coarse-grained nature.<n>Feature normalization suppresses discnative, domain-sensitive semantic details owing to its rigid design.
  arXiv  Detail & Related papers  (2026-08-09T16:43:04Z)
• Federated Joint Learning for Domain and Class Generalization [15.177261433209301]
  textbfFedDCG is a novel approach that addresses both class and domain generalization in federated learning settings.<n>Experiments show that textbfFedDCG outperforms state-of-the-art baselines in terms of accuracy and robustness.
  arXiv  Detail & Related papers  (2026-01-18T04:24:11Z)
• Text-Driven Causal Representation Learning for Source-Free Domain Generalization [82.75041792888274]
  We propose TDCRL, the first method to integrate causal inference into the source-free domain generalization setting.<n>Our approach offers a clear and effective mechanism to achieve robust, domain-invariant features, ensuring robust generalization.
  arXiv  Detail & Related papers  (2025-07-14T06:20:42Z)
• Prompt Disentanglement via Language Guidance and Representation Alignment for Domain Generalization [75.88719716002014]
  Domain Generalization (DG) seeks to develop a versatile model capable of performing effectively on unseen target domains.<n>Recent advances in pre-trained Visual Foundation Models (VFMs) have demonstrated considerable potential in enhancing the generalization capabilities of deep learning models.<n>We propose addressing this challenge by leveraging the controllable and flexible language prompt of the VFM.
  arXiv  Detail & Related papers  (2025-07-03T03:52:37Z)
• In the Era of Prompt Learning with Vision-Language Models [1.060608983034705]
  We introduce textscStyLIP, a novel domain-agnostic prompt learning strategy for Domain Generalization (DG) StyLIP disentangles visual style and content in CLIPs vision encoder by using style projectors to learn domain-specific prompt tokens. We also propose AD-CLIP for unsupervised domain adaptation (DA), leveraging CLIPs frozen vision backbone.
  arXiv  Detail & Related papers  (2024-11-07T17:31:21Z)
• StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization [85.18995948334592]
  Single domain generalization (single DG) aims at learning a robust model generalizable to unseen domains from only one training domain. State-of-the-art approaches have mostly relied on data augmentations, such as adversarial perturbation and style enhancement, to synthesize new data. We propose emphStyDeSty, which explicitly accounts for the alignment of the source and pseudo domains in the process of data augmentation.
  arXiv  Detail & Related papers  (2024-06-01T02:41:34Z)
• Instance-Aware Domain Generalization for Face Anti-Spoofing [42.36157210235893]
  Face anti-spoofing (FAS) has been recently studied to improve the generalization on unseen scenarios. Previous methods rely on domain labels to align the distribution of each domain for learning domain-invariant representations. We propose a novel perspective for DG FAS that aligns features on the instance level without the need for domain labels.
  arXiv  Detail & Related papers  (2023-04-12T06:41:16Z)
• Compound Domain Generalization via Meta-Knowledge Encoding [55.22920476224671]
  We introduce Style-induced Domain-specific Normalization (SDNorm) to re-normalize the multi-modal underlying distributions. We harness the prototype representations, the centroids of classes, to perform relational modeling in the embedding space. Experiments on four standard Domain Generalization benchmarks reveal that COMEN exceeds the state-of-the-art performance without the need of domain supervision.
  arXiv  Detail & Related papers  (2022-03-24T11:54:59Z)
• Structured Latent Embeddings for Recognizing Unseen Classes in Unseen Domains [108.11746235308046]
  We propose a novel approach that learns domain-agnostic structured latent embeddings by projecting images from different domains. Our experiments on the challenging DomainNet and DomainNet-LS benchmarks show the superiority of our approach over existing methods.
  arXiv  Detail & Related papers  (2021-07-12T17:57:46Z)
• Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation [95.10255219396109]
  We propose a fine-grained adversarial learning strategy for class-level feature alignment. We adopt a fine-grained domain discriminator that not only plays as a domain distinguisher, but also differentiates domains at class level. An analysis with Class Center Distance (CCD) validates that our fine-grained adversarial strategy achieves better class-level alignment.
  arXiv  Detail & Related papers  (2020-07-17T20:50:59Z)
• Contextual-Relation Consistent Domain Adaptation for Semantic Segmentation [44.19436340246248]
  This paper presents an innovative local contextual-relation consistent domain adaptation technique. It aims to achieve local-level consistencies during the global-level alignment. Experiments demonstrate its superior segmentation performance as compared with state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-05T19:00:46Z)

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.