CONTEXTOR: Contextualized High-order Contrastive Learning

Ze Cai, Hanzhe Liang, Sihang Zeng, Binbin Zhou, Jun Wen
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:10591-10609, 2026.

Abstract

High-order relations involving multiple interacting entities are commonly encountered, particularly in biomedical domains. Existing relational learning methods typically learn static entity representations and assume symmetric relation inference, which can be inadequate for capturing context-dependent entity functions and the inherent asymmetry of high-order relations. In this paper, we propose Contextualized High-order Contrastive Learning (CONTEXTOR), a general and plug-and-play framework that formulates high-order relation inference as a dynamic query–response process. Specifically, CONTEXTOR decomposes each high-order relation into multiple incomplete query tuples and their corresponding response entities. Given a query tuple, we contextualize candidate response entity representations via an asymmetric conditional modulation, and align queries with their corresponding contextualized responses through multi-fold contrastive learning. Extensive experiments on benchmark datasets spanning multiple biomedical tasks demonstrate that CONTEXTOR consistently outperforms existing methods across diverse evaluation settings. Code is available at https://github.com/ZJUDataIntelligence/CONTEXTOR.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-cai26f, title = {{CONTEXTOR}: Contextualized High-order Contrastive Learning}, author = {Cai, Ze and Liang, Hanzhe and Zeng, Sihang and Zhou, Binbin and Wen, Jun}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {10591--10609}, 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/cai26f/cai26f.pdf}, url = {https://proceedings.mlr.press/v306/cai26f.html}, abstract = {High-order relations involving multiple interacting entities are commonly encountered, particularly in biomedical domains. Existing relational learning methods typically learn static entity representations and assume symmetric relation inference, which can be inadequate for capturing context-dependent entity functions and the inherent asymmetry of high-order relations. In this paper, we propose Contextualized High-order Contrastive Learning (CONTEXTOR), a general and plug-and-play framework that formulates high-order relation inference as a dynamic query–response process. Specifically, CONTEXTOR decomposes each high-order relation into multiple incomplete query tuples and their corresponding response entities. Given a query tuple, we contextualize candidate response entity representations via an asymmetric conditional modulation, and align queries with their corresponding contextualized responses through multi-fold contrastive learning. Extensive experiments on benchmark datasets spanning multiple biomedical tasks demonstrate that CONTEXTOR consistently outperforms existing methods across diverse evaluation settings. Code is available at https://github.com/ZJUDataIntelligence/CONTEXTOR.} }
Endnote
%0 Conference Paper %T CONTEXTOR: Contextualized High-order Contrastive Learning %A Ze Cai %A Hanzhe Liang %A Sihang Zeng %A Binbin Zhou %A Jun Wen %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-cai26f %I PMLR %P 10591--10609 %U https://proceedings.mlr.press/v306/cai26f.html %V 306 %X High-order relations involving multiple interacting entities are commonly encountered, particularly in biomedical domains. Existing relational learning methods typically learn static entity representations and assume symmetric relation inference, which can be inadequate for capturing context-dependent entity functions and the inherent asymmetry of high-order relations. In this paper, we propose Contextualized High-order Contrastive Learning (CONTEXTOR), a general and plug-and-play framework that formulates high-order relation inference as a dynamic query–response process. Specifically, CONTEXTOR decomposes each high-order relation into multiple incomplete query tuples and their corresponding response entities. Given a query tuple, we contextualize candidate response entity representations via an asymmetric conditional modulation, and align queries with their corresponding contextualized responses through multi-fold contrastive learning. Extensive experiments on benchmark datasets spanning multiple biomedical tasks demonstrate that CONTEXTOR consistently outperforms existing methods across diverse evaluation settings. Code is available at https://github.com/ZJUDataIntelligence/CONTEXTOR.
APA
Cai, Z., Liang, H., Zeng, S., Zhou, B. & Wen, J.. (2026). CONTEXTOR: Contextualized High-order Contrastive Learning. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:10591-10609 Available from https://proceedings.mlr.press/v306/cai26f.html.

Related Material