Paper
OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework
Authors
Ben Chen, Siyuan Wang, Yufei Ma, Zihan Liang, Xuxin Zhang, Yue Lv, Ying Yang, Huangyu Dai, Lingtao Mao, Tong Zhao, Zhipeng Qian, Xinyu Sun, Zhixin Zhai, Yang Zhao, Bochao Liu, Jingshan Lv, Xiao Liang, Hui Kong, Jing Chen, Han Li, Chenyi Lei, Wenwu Ou, Kun Gai
Abstract
Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose \textbf{OneSearch-V2}, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\% item CTR, +3.05\% buyer conversion rate, and +2.11\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.65\% in page good rate and +1.37\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency.
Metadata
Related papers
Fractal universe and quantum gravity made simple
Fabio Briscese, Gianluca Calcagni • 2026-03-25
POLY-SIM: Polyglot Speaker Identification with Missing Modality Grand Challenge 2026 Evaluation Plan
Marta Moscati, Muhammad Saad Saeed, Marina Zanoni, Mubashir Noman, Rohan Kuma... • 2026-03-25
LensWalk: Agentic Video Understanding by Planning How You See in Videos
Keliang Li, Yansong Li, Hongze Shen, Mengdi Liu, Hong Chang, Shiguang Shan • 2026-03-25
Orientation Reconstruction of Proteins using Coulomb Explosions
Tomas André, Alfredo Bellisario, Nicusor Timneanu, Carl Caleman • 2026-03-25
The role of spatial context and multitask learning in the detection of organic and conventional farming systems based on Sentinel-2 time series
Jan Hemmerling, Marcel Schwieder, Philippe Rufin, Leon-Friedrich Thomas, Mire... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.24422v1</id>\n <title>OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework</title>\n <updated>2026-03-25T15:33:34Z</updated>\n <link href='https://arxiv.org/abs/2603.24422v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.24422v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose \\textbf{OneSearch-V2}, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\\% item CTR, +3.05\\% buyer conversion rate, and +2.11\\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.65\\% in page good rate and +1.37\\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.IR'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n <published>2026-03-25T15:33:34Z</published>\n <arxiv:comment>Key codes are available at https://github.com/benchen4395/onesearch-family. Feel free to contact benchen4395@gmail.com</arxiv:comment>\n <arxiv:primary_category term='cs.IR'/>\n <author>\n <name>Ben Chen</name>\n </author>\n <author>\n <name>Siyuan Wang</name>\n </author>\n <author>\n <name>Yufei Ma</name>\n </author>\n <author>\n <name>Zihan Liang</name>\n </author>\n <author>\n <name>Xuxin Zhang</name>\n </author>\n <author>\n <name>Yue Lv</name>\n </author>\n <author>\n <name>Ying Yang</name>\n </author>\n <author>\n <name>Huangyu Dai</name>\n </author>\n <author>\n <name>Lingtao Mao</name>\n </author>\n <author>\n <name>Tong Zhao</name>\n </author>\n <author>\n <name>Zhipeng Qian</name>\n </author>\n <author>\n <name>Xinyu Sun</name>\n </author>\n <author>\n <name>Zhixin Zhai</name>\n </author>\n <author>\n <name>Yang Zhao</name>\n </author>\n <author>\n <name>Bochao Liu</name>\n </author>\n <author>\n <name>Jingshan Lv</name>\n </author>\n <author>\n <name>Xiao Liang</name>\n </author>\n <author>\n <name>Hui Kong</name>\n </author>\n <author>\n <name>Jing Chen</name>\n </author>\n <author>\n <name>Han Li</name>\n </author>\n <author>\n <name>Chenyi Lei</name>\n </author>\n <author>\n <name>Wenwu Ou</name>\n </author>\n <author>\n <name>Kun Gai</name>\n </author>\n </entry>"
}