Deep Discriminative Structure Proxy Hashing for Cross-modal Retrieval

Kun Cheng, Qibing Qin, Lei Huang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:18997-19007, 2026.

Abstract

Existing proxy-based hashing methods optimize samples toward independently learned proxies using isolated similarity constraints. Although efficient, this design overlooks the fact that proxies are learned jointly but lack explicit relational or competitive interactions during optimization. Consequently, proxy responses to a sample are often accumulated rather than contrasted, leading to weakly defined decision regions and limited discriminative structure in the Hamming space. In contrast, our method organizes multiple proxies into sample-specific relational structures, enabling proxies to interact and compete when responding to each sample. Through structure-guided learning, these interactions explicitly contrast positive and negative proxy responses, thereby shaping clearer and more discriminative decision boundaries. Extensive experiments on standard cross-modal benchmarks demonstrate that this structured discrimination consistently improves retrieval accuracy and embedding separability. The source code is available at https://github.com/QinLab-WFU/DDSPH.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-cheng26k, title = {Deep Discriminative Structure Proxy Hashing for Cross-modal Retrieval}, author = {Cheng, Kun and Qin, Qibing and Huang, Lei}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {18997--19007}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/cheng26k/cheng26k.pdf}, url = {https://proceedings.mlr.press/v306/cheng26k.html}, abstract = {Existing proxy-based hashing methods optimize samples toward independently learned proxies using isolated similarity constraints. Although efficient, this design overlooks the fact that proxies are learned jointly but lack explicit relational or competitive interactions during optimization. Consequently, proxy responses to a sample are often accumulated rather than contrasted, leading to weakly defined decision regions and limited discriminative structure in the Hamming space. In contrast, our method organizes multiple proxies into sample-specific relational structures, enabling proxies to interact and compete when responding to each sample. Through structure-guided learning, these interactions explicitly contrast positive and negative proxy responses, thereby shaping clearer and more discriminative decision boundaries. Extensive experiments on standard cross-modal benchmarks demonstrate that this structured discrimination consistently improves retrieval accuracy and embedding separability. The source code is available at https://github.com/QinLab-WFU/DDSPH.} }
Endnote
%0 Conference Paper %T Deep Discriminative Structure Proxy Hashing for Cross-modal Retrieval %A Kun Cheng %A Qibing Qin %A Lei Huang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-cheng26k %I PMLR %P 18997--19007 %U https://proceedings.mlr.press/v306/cheng26k.html %V 306 %X Existing proxy-based hashing methods optimize samples toward independently learned proxies using isolated similarity constraints. Although efficient, this design overlooks the fact that proxies are learned jointly but lack explicit relational or competitive interactions during optimization. Consequently, proxy responses to a sample are often accumulated rather than contrasted, leading to weakly defined decision regions and limited discriminative structure in the Hamming space. In contrast, our method organizes multiple proxies into sample-specific relational structures, enabling proxies to interact and compete when responding to each sample. Through structure-guided learning, these interactions explicitly contrast positive and negative proxy responses, thereby shaping clearer and more discriminative decision boundaries. Extensive experiments on standard cross-modal benchmarks demonstrate that this structured discrimination consistently improves retrieval accuracy and embedding separability. The source code is available at https://github.com/QinLab-WFU/DDSPH.
APA
Cheng, K., Qin, Q. & Huang, L.. (2026). Deep Discriminative Structure Proxy Hashing for Cross-modal Retrieval. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:18997-19007 Available from https://proceedings.mlr.press/v306/cheng26k.html.

Related Material