OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search

Ben Chen, Xian Guo, Siyuan Wang, Zihan Liang, Yufei Ma, Yue Lv, Chenyi Lei, Yuqing Ding, Wenwu Ou, Han Li, Kun Gai
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:14392-14415, 2026.

Abstract

Traditional e-commerce search systems employ multi-stage cascading architectures that suffer from fragmented computation and optimization objective collisions across stages, ultimately limiting their performance ceiling. We propose OneSearch, the first industrial-deployed end-to-end generative framework for e-commerce search, featuring three key innovations: (1) Keyword-enhanced Hierarchical Quantization Encoding to preserve hierarchical semantics and distinctive item attributes while maintaining strong query-item relevance constraints; (2) multi-view user behavior sequence injection that constructs behavior-driven user IDs and incorporates both explicit short-term and implicit long-term sequences; and (3) a Preference-Aware Reward System with multi-stage supervised fine-tuning and adaptive reward-weighted ranking to capture fine-grained user preferences. Extensive offline evaluations demonstrate its superior performance, while online A/B tests achieve statistically significant improvements: +1.67% item CTR, +2.40% buyer, and +3.22% order volume. OneSearch reduces operational expenditure by 75.40%, improves Model FLOPs Utilization from 3.26% to 27.32%, and has been successfully deployed across multiple search scenarios in Kuaishou, serving millions of users daily. Code is in https://github.com/benchen4395/onesearch-family.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26al, title = {{O}ne{S}earch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search}, author = {Chen, Ben and Guo, Xian and Wang, Siyuan and Liang, Zihan and Ma, Yufei and Lv, Yue and Lei, Chenyi and Ding, Yuqing and Ou, Wenwu and Li, Han and Gai, Kun}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {14392--14415}, 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/chen26al/chen26al.pdf}, url = {https://proceedings.mlr.press/v306/chen26al.html}, abstract = {Traditional e-commerce search systems employ multi-stage cascading architectures that suffer from fragmented computation and optimization objective collisions across stages, ultimately limiting their performance ceiling. We propose OneSearch, the first industrial-deployed end-to-end generative framework for e-commerce search, featuring three key innovations: (1) Keyword-enhanced Hierarchical Quantization Encoding to preserve hierarchical semantics and distinctive item attributes while maintaining strong query-item relevance constraints; (2) multi-view user behavior sequence injection that constructs behavior-driven user IDs and incorporates both explicit short-term and implicit long-term sequences; and (3) a Preference-Aware Reward System with multi-stage supervised fine-tuning and adaptive reward-weighted ranking to capture fine-grained user preferences. Extensive offline evaluations demonstrate its superior performance, while online A/B tests achieve statistically significant improvements: +1.67% item CTR, +2.40% buyer, and +3.22% order volume. OneSearch reduces operational expenditure by 75.40%, improves Model FLOPs Utilization from 3.26% to 27.32%, and has been successfully deployed across multiple search scenarios in Kuaishou, serving millions of users daily. Code is in https://github.com/benchen4395/onesearch-family.} }
Endnote
%0 Conference Paper %T OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search %A Ben Chen %A Xian Guo %A Siyuan Wang %A Zihan Liang %A Yufei Ma %A Yue Lv %A Chenyi Lei %A Yuqing Ding %A Wenwu Ou %A Han Li %A Kun Gai %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-chen26al %I PMLR %P 14392--14415 %U https://proceedings.mlr.press/v306/chen26al.html %V 306 %X Traditional e-commerce search systems employ multi-stage cascading architectures that suffer from fragmented computation and optimization objective collisions across stages, ultimately limiting their performance ceiling. We propose OneSearch, the first industrial-deployed end-to-end generative framework for e-commerce search, featuring three key innovations: (1) Keyword-enhanced Hierarchical Quantization Encoding to preserve hierarchical semantics and distinctive item attributes while maintaining strong query-item relevance constraints; (2) multi-view user behavior sequence injection that constructs behavior-driven user IDs and incorporates both explicit short-term and implicit long-term sequences; and (3) a Preference-Aware Reward System with multi-stage supervised fine-tuning and adaptive reward-weighted ranking to capture fine-grained user preferences. Extensive offline evaluations demonstrate its superior performance, while online A/B tests achieve statistically significant improvements: +1.67% item CTR, +2.40% buyer, and +3.22% order volume. OneSearch reduces operational expenditure by 75.40%, improves Model FLOPs Utilization from 3.26% to 27.32%, and has been successfully deployed across multiple search scenarios in Kuaishou, serving millions of users daily. Code is in https://github.com/benchen4395/onesearch-family.
APA
Chen, B., Guo, X., Wang, S., Liang, Z., Ma, Y., Lv, Y., Lei, C., Ding, Y., Ou, W., Li, H. & Gai, K.. (2026). OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:14392-14415 Available from https://proceedings.mlr.press/v306/chen26al.html.

Related Material