Conference Open access

SPACE: Semantic Projection and Alignment of Clip Embeddings for Domain Adaptation

J. R. Manesco D. Jodas Douglas Rodrigues Leandro A. Passos J. P. Papa

Sep 2026 · International Conference on Information Photonics · pp. 1-6 · 0 citations · 19 references

Computer Science

TL;DR

SPACE is introduced, a method that exploits the semantic structure of CLIP's vision-language space for domain adaptation by applying Singular Value Decomposition to CLIP embeddings of class descriptions, yielding an orthogonal basis that captures semantic relationships among categories.

Abstract

A fundamental challenge in deploying vision models is domain shift, which arises when training and test data follow different distributions, leading to degraded performance. This challenge is amplified when the same semantic concept appears under distinct visual forms, such as photographs and sketches, where visual similarity is weak despite semantic correspondence. Existing unsupervised domain-adaptation methods aim to align distributions across domains but often ignore semantic relationships among samples of the same class. To address this issue, this paper introduces SPACE, a method that exploits the semantic structure of CLIP's vision-language space for domain adaptation. The key idea is to use text descriptions as semantic anchors by applying Singular Value Decomposition to CLIP embeddings of class descriptions, yielding an orthogonal basis that captures semantic relationships among categories. Visual features from both domains are projected into this semantic subspace, aligning images based on meaning rather than appearance.

DOI arXiv Semantic Scholar

Read PDF

Similar papers

Preprint Sep 2026

Generative Uncertainty as a Self-supervised Signal for Semantic Similarity Learning

A novel self-supervised approach that leverages generative uncertainty from text-to-video (T2V) diffusion models to learn semantic similarity without human annotations to identify stable semantic features within existing pretrained representations, such as VideoMAE and V-JEPA.

Enrico Pallotta, Sina Raoufi, Lars Doorenbos et al. · 0 citations

#machine learning Preprint Sep 2026

Spherical Interpolation for Backward-Compatible Multimodal Representations

Contrastive vision-language models map visual and textual representations into a shared normalized embedding space, making cosine similarity the natural metric for cross-modal retrieval. A practical challenge arises during model upgrades: independently trained models generally produce incompatible representation spaces...

Simone Ricci, Niccoló Biondi, F. Pernici · 0 citations

#artificial intelligence Preprint Oct 2026

Revisiting Visual Representation Enhancement of VLMs via Kernel Canonical Correlation Analysis

Vision-language models such as CLIP exhibit strong semantic generalization, but remain limited in fine-grained visual perception. A recent work named KUEA presents a natural remedy by finetuning the image encoder under the supervision of the vision-centric DINOv2 to align their kernel matrices element-wisely, while reg...

Pei-Lin Yang, Xiao-Yu Liu, Jian Sun et al. · 0 citations

Aug 2026

Semantic manifold-aware cross-modal similarity learning.

A Semantic Manifold-Aware Similarity Learning (SMSL) framework, where the term "semantic manifold" is used in an operational sense to denote a topology-aware organization of textual semantic units induced from token/entity interactions, rather than a strict low-dimensional differentiable manifold.

Li-Qi Zhu, De-Zhi Han, Chong-Qing Chen · Neural Networks · 0 citations

Preprint Aug 2026

OPAL: Orthonormal Prototype Alignment Learning for Interpretable Image Classification

This work introduces Orthonormal Prototype Alignment Learning (OPAL), a single-stage, end-to-end framework that simplifies interpretable classification by framing classification as a direct representation alignment task, and eliminates the need for auxiliary losses.

I. Carretero, G. Angulo, R. del Amor et al. · 0 citations

Oct 2026

Prototype Matters: Domain-Wise Prototype Induction for Zero-Shot Recognition.

Zero-Shot Learning (ZSL) has achieved remarkable progress, yet most existing methods still rely heavily on predefined semantic embeddings as supervisory signals. Such semantics are often incomplete, inaccurate, and insufficiently discriminative, resulting in ambiguous decision boundaries and limited generalization to u...

Bo-Cheng Ren, Yuan-Yuan Yi, Qing-Chen Zhang et al. · IEEE Transactions on Pattern... · 0 citations