Attribute-Prompted Kernel Hashing for Unsupervised Data-Efficient Cross-Modal Retrieval

• URL: http://arxiv.org/abs/2607.00379v1
• Date: Wed, 01 Jul 2026 03:23:52 GMT
• Title: Attribute-Prompted Kernel Hashing for Unsupervised Data-Efficient Cross-Modal Retrieval
• Abstract summary: Cross-modal hashing enables efficient retrieval of semantically related instances across different modalities.<n>Existing unsupervised methods rely heavily on large-scale image-text pairs.<n>We propose Attribute-Prompted Kernel Hashing (APKH), a novel data-efficient approach that constructs a modality-aligned Hamming space.
• Score: 33.22221387810735
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Unsupervised cross-modal hashing enables efficient retrieval of semantically related instances across different modalities without requiring manual semantic annotation. However, existing unsupervised methods rely heavily on large-scale image-text pairs. Collecting such data can be costly, particularly in scenarios where well-aligned pairs are scarce due to privacy and specialized constraints. More critically, existing methods tend to overfit to seen training data, restricting their generalization performance on unseen categories that the constrained training data cannot cover. To address these limitations, we propose Attribute-Prompted Kernel Hashing (APKH), a novel data-efficient approach that constructs a compact, modality-aligned Hamming space driven by the generalized attribute priors of vision-language foundation models. Specifically, APKH introduces two core modules: Context-optimized Attribute Kernel Mapping (CAKM) and Kernel-Smoothed Contrastive Alignment (KSCA). CAKM formulates cross-modal alignment through hyperspherical Radial Basis Function kernel mapping, optimizing dynamic attribute kernels via prompt learning to capture modality-invariant semantics. Furthermore, KSCA extends conventional point-to-point contrastive learning by modeling limited paired data as continuous kernel distributions. This explicit smoothing of the modality gap alleviates overfitting to sparse pairwise correlations. Extensive experiments demonstrate that APKH outperforms state-of-the-art hashing methods in the challenging cross-modal retrieval tasks from seen to unseen categories under data-constrained scenarios.

Related papers

• Adapformer: Adaptive Channel Management for Multivariate Time Series Forecasting [49.40321003932633]
  Adapformer is an advanced Transformer-based framework that merges the benefits of CI and CD methodologies through effective channel management.<n>Adapformer achieves superior performance over existing models, enhancing both predictive accuracy and computational efficiency.
  arXiv  Detail & Related papers  (2025-11-18T16:24:05Z)
• Localized Kernel Projection Outlyingness: A Two-Stage Approach for Multi-Modal Outlier Detection [0.0]
  Two-Stage LKPLO is a novel multi-stage outlier detection framework.<n>It overcomes the coexisting limitations of conventional projection-based methods.<n>It achieves state-of-the-art performance on challenging datasets.
  arXiv  Detail & Related papers  (2025-10-28T03:53:46Z)
• Instance-Aware Robust Consistency Regularization for Semi-Supervised Nuclei Instance Segmentation [53.94176748542936]
  We propose an Instance-Aware Robust Consistency Regularization Network (IRCR-Net) for accurate instance-level nuclei segmentation.<n>We incorporate morphological prior knowledge of nuclei in pathological images and utilize these priors to assess the quality of pseudo-labels generated from unlabeled data.
  arXiv  Detail & Related papers  (2025-10-10T12:32:32Z)
• Hierarchical Self-Supervised Representation Learning for Depression Detection from Speech [51.14752758616364]
  Speech-based depression detection (SDD) is a promising, non-invasive alternative to traditional clinical assessments.<n>We propose HAREN-CTC, a novel architecture that integrates multi-layer SSL features using cross-attention within a multitask learning framework.<n>The model achieves state-of-the-art macro F1-scores of 0.81 on DAIC-WOZ and 0.82 on MODMA, outperforming prior methods across both evaluation scenarios.
  arXiv  Detail & Related papers  (2025-10-05T09:32:12Z)
• Cross-Modal Retrieval with Cauchy-Schwarz Divergence [26.94915416778522]
  Cross-modal retrieval requires robust alignment of heterogeneous data types.<n>Most existing methods rely on distributional alignment techniques such as Kullback-Leibler divergence.<n>We introduce the Cauchy-Schwarz (CS) divergence, a hyper parameter-free measure that improves both training stability and retrieval performance.
  arXiv  Detail & Related papers  (2025-09-15T08:55:15Z)
• Disentangled Noisy Correspondence Learning [56.06801962154915]
  Cross-modal retrieval is crucial in understanding latent correspondences across modalities. DisNCL is a novel information-theoretic framework for feature Disentanglement in Noisy Correspondence Learning.
  arXiv  Detail & Related papers  (2024-08-10T09:49:55Z)
• Kernel Correlation-Dissimilarity for Multiple Kernel k-Means Clustering [21.685153346752124]
  Current methods enhance information diversity and reduce redundancy by exploiting interdependencies among multiple kernels based on correlations or dissimilarities. We introduce a novel method that systematically integrates both kernel correlation and dissimilarity. By emphasizing the coherence between kernel correlation and dissimilarity, our method offers a more objective and transparent strategy for extracting non-linear information.
  arXiv  Detail & Related papers  (2024-03-06T04:24:43Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)
• Joint Embedding Self-Supervised Learning in the Kernel Regime [21.80241600638596]
  Self-supervised learning (SSL) produces useful representations of data without access to any labels for classifying the data. We extend this framework to incorporate algorithms based on kernel methods where embeddings are constructed by linear maps acting on the feature space of a kernel. We analyze our kernel model on small datasets to identify common features of self-supervised learning algorithms and gain theoretical insights into their performance on downstream tasks.
  arXiv  Detail & Related papers  (2022-09-29T15:53:19Z)
• Towards Uncovering the Intrinsic Data Structures for Unsupervised Domain Adaptation using Structurally Regularized Deep Clustering [119.88565565454378]
  Unsupervised domain adaptation (UDA) is to learn classification models that make predictions for unlabeled data on a target domain. We propose a hybrid model of Structurally Regularized Deep Clustering, which integrates the regularized discriminative clustering of target data with a generative one. Our proposed H-SRDC outperforms all the existing methods under both the inductive and transductive settings.
  arXiv  Detail & Related papers  (2020-12-08T08:52:00Z)

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.